(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Synthesis and biological activity of isoquinoline analogues of streptonigrin"

SYNTHESIS AND BIOLOGICAL ACTIVITY OF ISOQUINOLINE 
ANALOGUES OF STREPTONIGRIN 



By 

JOSEPH WARREN BEACH 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 
OF THE UNIVERSITY OF FLORIDA IN 
PARTIAL FULFILLMENT OF THE REQUIREMENTS 
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY 



UNIVERSITY OF FLORIDA 
1987 



To the Memory of My Mother 
Dorothy Bailey Beach 



ACKNOWLEDGMENTS 

I acknowledge my great appreciation to Dr. K. V. Rao for his 
guidance and supervision during the course of this work and to the 
other members of my supervisory committee, Dr. K. B. Sloan and Dr. J. 
A. Zoltewicz. I would also like to thank Dr. John Perrin and Dr. 
Richard H. Hammer for their assistance and encouragement. 

I would also like to thank Greg Wheeler for his help in setting 
up the graphs. 



ill 



TABLE OF CONTENTS 

Page 

ACKNOWLEDGMENTS 111 

LIST OF TABLES vi 

LIST OF FIGURES vii 

LIST OF ABBREVIATIONS x 

ABSTRACT xi 

CHAPTER 

I. INTRODUCTION 1 

Introduction 1 

Prior Synthetic Work 

II. THE SYNTHESIS OF 1-PHENYL AND l-( 2-PYRIDYL) -6- 

METHOXY- 7-AMIN0IS0QUIN0LINE-5.8-DI0NE 23 

Structure-Activity Relationships 23 

Other Heterocyclic and Carbocyclic Qui nones. 26 

Synthetic Approach to the Isoquinolines. . . 31 

Experimental 36 

III. SYNTHESIS OF NITRO AND AMINOPHENYLISOQUINOLINE 

DERIVATIVES 54 

Introduction 54 

Experimental 59 

IV. SYNTHESIS OF HOMO ANALOGUES: 1-BENZYL, 1-(2'NITR0)- 
BENZYL, AND l-(2' AMINO) BENZYL-6-METHOXY-7- 
AMIN0IS0QUIN0LINE-5.8-DI0NES 82 

Introduction 

Experimental 82 

V. METHYLAMINO AND DIMETHYLAMINO QUINONES 95 

Introduction 95 

Experimental 101 

iv 



Page 

VI. RESULTS AND DISCUSSION Ill 

Introduction Ill 

Materials and Methods 112 

Results 115 

Conclusions 130 

Further Research 131 

REFERENCES 132 

BIOGRAPHICAL SKETCH 136 



LIST OF TABLES 

Table Page 

I. ROOT INHIBITION ASSAY 121 

II. ANTIBIOTIC ASSAY 129 



VI 



LIST OF FIGURES 

Figure P fl 9 e 

1.1 Streptonigrin 

1.2 Proposed minimum structure for reverse tran- 

scriptase activity 8 

1.3 Synthesis of quino1ine-5,8-dione 9 

1.4 Synthesis of 2(2'-pyridyl )quinoline-5,8-dione .... 11 

1.5 Synthesis of 2(2'-pyridyl )quinoline-5,8-dione .... 12 

1.6 Synthesis directed toward the C-D ring 14 

1.7 Synthesis of the C-D ring of streptonigrin 15 

1.8 First total synthesis of streptonigrin 16 

1.9 Second total synthesis of streptonigrin 19 

1.10 Third total synthesis of streptonigrin 21 

1.11 Synthesis of the C-ring precursor 22 

2.1 Quinoline qui nones, relative potency (RP) to strepto- 

nigrin 24 

2.2 Some fused heterocyclic qui nones 27 

2.3 Comparison of quinolinedione to isoquinolinedione . . 29 

2.4 Some isoquinoline quinones synthesized previously . . 30 

2.5 l-Phenyl-6-methoxy-7-aminoisoquinoline-5,8-dione 

( 2.11 ) and l(2-pyridyl )-6-methoxy-7-aminoiso- 
quinoline-5,8-dione ( 2.17 ) 32 

2.6 Disconnection of l-(2-pyridyl )-6-methoxy-7-amino- 

isoquinoline-5,8-dione 33 

2.7 Synthesis of 1-phenyl- and l-(2-pyridyl )-6-methoxy- 

7-aminoisoqunoline-5,8-dione 35 

2.8 NMR spectrum of compound 2.1 39 

2.9 NMR spectrum of compound 2^_2 41 

vii 



Figure Page 

2.10 NMR spectrum of compound ^3 42 

2.11 NMR spectrum of compound 2^4 44 

2.12 NMR spectrum of compound 2.11 48 

2.13 NMR spectrum of compound 2.17 53 

3.1 Favorable spatial arrangement of a quinoline 

qui none compared to isoquinoline qui none. ... 55 

3.2 Possible favorable spatial arrangement of the 

isoquinolines 57 

3.3 Synthesis of amino- and nitro phenyl compounds. . . 58 

3.4 Structure of nitro- and aminophenyl isoquinoline 

ami noqui nones 60 

3.5 Reduction of nitro group by the borohydride .... 61 

3.6 NMR spectrum of compound 3^6 65 

3.7 NMR spectrum of compound 3^ 67 

3.8 NMR spectrum of compound 3J3 69 

3.9 NMR spectrum of compound 3.15 72 

3.10 NMR spectrum of compound 3.16 74 

3.11 NMR spectrum of compound 3.17 75 

3.12 NMR spectrum of compound 3.24 79 

3.13 NMR spectrum of compound 3.25 81 

4.1 Synthesis of benzyl derivatives 83 

4.2 Structure of benzyl, 2'nitrobenzyl and 2 amino- 

benzyl isoquinoline ami noqui nones 84 

4.3 NMR spectrum of compound 4^ 89 

4.4 NMR spectrum of compound 4.13 92 

4.5 NMR spectrum of compound 4.14 94 

5.1 Comparison of 6-methoxy-7-aminoquinone with 

6-methyl ami no-7-bromoqui none 97 

viii 



Figure Page 

5.2 NMR spectrum of compound 5^1 98 

5.3 Unusual reactions 99 

5.4 IR spectrum of compound 5^2 100 

5.5 Structure of compounds synthesized 102 

5.6 NMR spectrum of compound 5^_3 105 

5.7 NMR spectrum of compound bA_ 106 

5.8 NMR spectrum of compound 5_^5_ 107 

5.9 NMR spectrum of compound 5^6 109 

5.10 NMR spectrum of compound 5^7 110 

6.1 Structures of compounds tested in the root growth 

inhibit assay and antibiotic assay 113 

6.2 Comparison of compounds 2.11 , 2.17 and strepto- 

nigrin (SN) 116 

6.3 Comparison of compounds 3.7 , 3.16 , 3.25 and 

streptonigrin (SN) 117 

6.4 Comparison of compounds 3.6 , 3.15 , 3.24 and 

streptonigrin (SN) 118 

6.5 Comparison of compounds 2.11 , 4.6 and strepto- 

nigrin (SN) 119 

6.6 Comparison of 2.17 , streptonigrin destrioxyphenyl 

streptonigrin (dSN) and pyridylquinoline 

ami noqui none (isoPyQ) 120 

6.7 Comparison of compound 2.11 , streptonigrin (SN) 

pyridylquinoline ami noqui no! ine (isoPyQ) .... 124 

6.8 Comparison of compounds 2.11 , 4.6 and strepto- 

nigrin (SN) 125 

6.9 Comparison of compounds 3.7 , 3.16 , 3.25 , and 

streptonigrin (SN) I 26 

6.10 Comparison of compounds 3.15 , 3.24 and strepto- 

nigrin (SN) 127 

6.11 Comparison of compounds 4.6 , 5.3 and strepto- 

nigrin (SN) 128 

ix 



LIST OF ABBREVIATIONS 

ATP adenosine triphosphate 

CDC1 3 deuteriochloroform 

dec decomposed 

DMF N,N-dimethylformamide 

DMSOdg hexadeuteriodimethyl sulfoxide 

DNA deoxyribonucleic acid 

eq equivalent 

g gram 

HTLV III human t-lymphotropic virus, type 3 

MCPBA m-chloroperbenzoic acid 

mg milligram 

NADH nictoinamide adenine di nucleotide (reduced form) 

NADPH nicotinamide adenine dinucleotide phosphate (reduced 
form) 

nM nanomole 

POCI3 phosphorus oxychloride 

ppm parts per million 

RNA ribonucleic acid 

THF tetrahydrofuran 

T m DNA melt temperature 

TLC thin layer chromatography 

yg microgram 



Abstract of Dissertation Presented to the Graduate School 

of the University of Florida in Partial Fulfillment of the 

Requirements for the Degree of Doctor of Philosophy 

SYNTHESIS AND BIOLOGICAL ACTIVITY OF ISOQUINOLINE 
ANALOGUES OF STREPTONIGRIN 

By 

Joseph Warren Beach 

December 1987 

Chairman: Dr. Koppaka V. Rao 

Major Department: Medicinal Chemistry 

Streptonigrin, an antitumor antibiotic isolated from Strepto - 
myces flocculus , is one of the most potent antitumor compounds known. 
It shows activity against a wide variety of experimental animal tu- 
mors as well as several viruses, including herpes simplex I and HTLV 
III. Its clinical effectiveness was also demonstrated as a single 
drug or in combination. In addition, the compound is a potent plant 
growth inhibitor. 

Streptonigrin has a unique phenyl -pyridyl qui noline quinone 
system with a number of structural elements capable of complexation 
with divalent ions and this ability serves to play an important role 
in its activity. A number of structural variations have been made to 
understand the requirements for activity and most of these were made 
in rings A, C and D. However, variations of ring B, such as 
replacement of the qui noline ring by an isoqui noline ring, have not 
been studied. The present objective was to synthesize a number of 

xi 



"isoquinoline analogues" and examine their activity using a microbio- 
logical assay (B_. subtil i s ) and a root growth inhibition assay (cress 
seedlings). 

As a first stage, l-phenyl-6-methoxy-7-aminoisoquinoline-5,8- 
dione and l-(2-pyridyl )-6-methoxy-7-aminoisoquinoline-5,8-dione were 
synthesized using a 10-step sequence. These compounds were found to 
be equipotent with streptonigrin in the root growth inhibition assay 
but to be much less potent than streptonigrin in the antibacterial 
assay. 

In order to study the effect of the amine function in ring C, 
the o- m- and p-aminophenylisoquinolines were synthesized via the 
corresponding nitro derivatives. Similarly, to study the effect of 
conjugation in the system, a group of 1-benzyl isoquinoline quinones 
was also prepared. When tested in the root growth inhibition assay, 
the 1-benzyl and 1-phenyl isoquinoline derivatives were found to be 
equipotent. However, the ami nophenyl isoquinoline quinones were less 
potent than the phenyl isoquinolines,- whereas the corresponding nitro- 
phenylisoquinolines were of greater potency than the phenyl isoquino- 
line. The implications of the structures to metal complexation and 
hydrogen bonding were discussed. 



xn 



CHAPTER I 
INTRODUCTION 

Introduction 

Streptonigrin (Figure 1.1), an antitumor antibiotic isolated by 
Rao and Cull en in 1960, from Streptomyces flocculus (1), is one of 
the most active compounds known for the treatment of human cancers 
(2). Early investigations showed that streptonigrin possessed activ- 
ity against a wide range of experimental tumors in vivo (3, 4), such 
as sarcoma 180, Lewis lung carcinoma, Ridgeway and Wagner osteogenic 
sarcomas, Walker 256 carcinosarcoma, mammary adenocarcinoma, Flexner- 
Jobling carcinoma and Iglesias ovarian tumor. Streptonigrin also 
possesses activity against several viruses (5), including herpes sim- 
plex I and HTLV III (6). The potential clinical use of streptonigrin 
has been hampered by early reports of its severe toxicity (7, 8); 
however, these early studies were conducted in terminal or advanced 
cancer patients who had previously received radiation therapy or 
other bone marrow depressing drugs. Many antitumor drugs cause bone 
marrow depression at doses lower than normal , when given soon after 
radiation therapy (9). 

Comparative studies conducted using orally administered chlor- 
ambucil and streptonigrin against chronic lymphocytic leukemia (10) 
and malignant lymphoma (11) indicated that these two drugs were 
equally effective in the treatment of these disease states. Strep- 
tonigrin showed only slightly more toxicity, usually manifested as 
thrombocytopenia, a decrease in the platelet count, and mild to 
moderate gastrointestinal toxicity which could be due to oral 

1 



6 CH 



^vCOOH 2 




Figure 1.1 Streptonigrin. 



administration. Later, streptonigrin therapy in a group of patients 
with lymphomas refractory to conventional therapy produced approxi- 
mately 36% complete and partial remissions (12). Due to this find- 
ing, streptonigrin has been incorporated into a number of experiment- 
al combination therapies (13-15). 

In one such trial (14), the effectiveness of the combination of 
vincristine and prednisone with cyclophosphamide or streptonigrin 
against non-Hodgkins lymphomas was investigated. The inclusion of 
streptonigrin in the therapy produced a higher percentage of complete 
remissions and less toxicity than the cyclophosphamide combination. 
This and other recent studies indicate that the early reports of the 
severe toxicity of streptonigrin were not due to an inherent toxicity 
of the drug but possibly due to the dose given or the patients used 
in the studies. 

The mechanism of action of streptonigrin is still not fully 
understood. It has been shown that streptonigrin does not intercal- 
ate into DNA as the anthracycline antibiotics or actinomycins do, as 
evidenced by a lack of change in the viscosity of DNA solutions when 
streptonigrin is added (16). It also does not crosslink DNA as mito- 
mycin C or N-mustards do, as shown by T m experiments (17) on DNA. 
Streptonigrin does, however, bind to DNA with the participation of 
divalent metal ions such as Zn (II), Cu (II), Mn (II), or Fe (II) 
(18, 19). Two different types of binding to DNA have been observed, 
one which is unstable to dialysis and the other which is stable and 
irreversible. Preferential binding to cytosine has been suggested 
(20, 21). 



4 

Streptonigrin seems to exert its antitumor effect by two mech- 
anisms (22). First, at low concentrations, by interfering with cell- 
ular electron transport and respiration, it causes depletion of NADH 
and NADPH, uncoupling of oxidative phosphorylation and a decrease in 
ATP production. Secondly, at higher concentrations, it disrupts 
cellular replication by causing extensive chromosomal breaks. Strep- 
tonigrin does show cell-cycle specificity and is preferentially bound 
to DNA during S phase (21). 

The cellular effects of streptonigrin resemble those of mito- 
mycin C, in that both cause extensive DNA degradation. However, 
streptonigrin does not cause interstrand crossl inking as mitomycin C 

does (20). 

For streptonigrin to express its full lethal effect, an elec- 
tron source, such as NADH or a reducing enzyme, oxygen and a divalent 
metal ion such as Fe (II) are required. In the absence of any one of 
these three, the activity is abolished or substantially reduced 

(19). 

NADH, in a one electron transfer, reduces streptonigrin to the 
semiquinone which undergoes rapid auto-oxidation back to the quinone. 
The free electron is transferred to oxygen producing a superoxide 
anion (Og 1 ). Two molecules of superoxide anion react with two pro- 
tons to produce hydrogen peroxide and oxygen. Superoxide anion may 

• 

also react with hydrogen peroxide to produce hydroxyl radical, OH 
(20). It has been proposed that OH is the ultimate reactive oxygen 
species which causes single strand breaks in DNA, thus inhibiting 



DNA, RNA and eventually, protein synthesis, and finally leading to 
cell death (20). By complexing with DNA in the presence of a metal 
ion, this lethal reaction takes place in close proximity to DNA. 
Hydrogen peroxide and superoxide anion are without effect on DNA, in 
contrast to hydroxyl radical which is highly reactive and short-lived 
and these properties would therefore require the radical to be gener- 
ated in close proximity to DNA (20, 23). 

Although the nature of the streptonigrin metal complex is not 
fully understood, Cu (II) and Fe (II) are known to accelerate strep- 
toni grin-induced single strand breaks in DNA, and Co (II) has been 
shown to be inhibitory in the same process (24). The bipyridyl 
structure of streptonigrin can bind to divalent ions such as Zn (II), 
Mn (II), Cu (II), Fe (II), and Co (II). Spectroscopic, kinetic and 
electrochemical redox properties indicate that the zinc and copper 
complexes of streptonigrin are structurally different. Alternative 
binding sites, other than the bipyridyl structure, may include the 
amine or the carboxylic acid group of the picolinic acid moiety (C 
ring) (25). It is also possible that metal complexation affects the 
redox properties of the aminoquinone moiety. Coordination of the 
qui none oxygen by an activating metal such as Cu (II) may account for 
the enhanced rate of chemical reduction via stabilization of the 
developing negative charge. The inhibitory effects toward strepto- 
nigrin reduction may be explained in terms of a metal ion-assisted 
tautomeric shift from the p-quinone to an o-azaquinone (25). Al- 
though the Zn (II) complex of streptonigrin is less susceptible to 
reduction, recent evidence suggests it may enhance DNA binding (26). 



In a recent study by Inouye et al . (6), where 150 antibiotics 
and their derivatives were tested, streptonigrin and its C2' carbox- 
amide derivatives (27) were found to be some of the most potent 
inhibitors of the avian myeloblastosis virus (a retrovirus), grown in 
cell culture. Cell death, as well as inhibition of viral replica- 
tion, was measured so that it could be determined whether the com- 
pounds acted directly on the virus or were active due to host cell 
death. The carboxamide derivatives of streptonigrin showed moderate 
cell toxicity while showing high viral inhibition, whereas strepto- 
nigrin showed high activity in both criteria. In another study (28), 
these same workers found that streptonigrin and its amide derivatives 
interfered directly with reverse transcriptase in a noncompetetive 
manner. It was suggested that this inhibition was an oxidation- 
reduction process, the enzyme being either oxidized or reduced to an 
inactive form by streptonigrin and its derivatives. From this 
information these workers concluded that a simplified quinoline quin- 
one would also be active and proposed structure 1.1 as a minimum 
structure (29). 

In the most recent work by these investigators (30, 31), a ser- 
ies of simple heterocyclic and carbocyclic quinones (quinoline, iso- 
quinoline, indole, and naphthoquinone) were tested as inhibitors of 
the retrovirus, avian myeloblastosis virus. The quinoline and iso- 
quinoline quinones were equi potent or more potent than the amide 
derivatives of streptonigrin whereas naphthoquinone and indol equi none 
were weaker inhibitors. The quinoline and isoquinoline quinones had 






similar substitution on the quinone ring as structure 1.1 (Figure 
1.2) and had either no or only simple (CH 3 or C=N) substitution on 
the nonquinone ring. The most potent compounds possessed an 
o-quinoid structure. 

These studies open the door for the possibility that other 
heterocyclic qui nones may have activity against the retrovirus HTLV 

III. 

Prior Synthetic Work 

The unique and highly functional!' zed nature of streptonigrin 
has presented an interesting challenge to the synthetic chemist over 
the past 20 years. These efforts have culminated in three total 
syntheses (32-34) and several syntheses of partial structural 
analogues of streptonigrin (24, 35, 36, 37). The structure, for the 
purpose of discussion may be divided into two parts, the A-B or 
quinoline quinone ring system and the C-D or pyridine-phenyl ring 
system. The synthetic routes to each will be discussed in turn. 

The quinoline quinone (A-B) system has been synthesized by two 
different routes. The first, used by Liao et al . (38, 39) (Figure 
1.3) involved the synthesis of 8-nitro-6-methoxyquinoline from 
2-nitro-p-anisidine via a Skraup synthesis, which is based on the 
condensation of an aniline with an a,e-unsaturated aldehyde. The 8- 
nitroquinoline was further nitrated, followed by reduction to the 
5,8-diaminoquinoline. Oxidation with sodium dichromate, followed by 
bromi nation gave the 6-methoxy-7-bromoquinoline-5,8-dione. Reaction 
of this with ammonia did not give the desired 7-aminoquinone but 




Figure 1.2. Proposed minimum structure for reverse 
transcriptase activity. Structure 1.1. 



HNO. 





C"] y Na 2 Cr 2°7 




Br 2 
AcOH 



-> 




NoN. 



CH 



-> 




Scheme I 



Figure 1.3. Synthesis of qirinoline-5,8-dione. 



10 

indeed, the 6-aminoquinone, resulting from direct displacement of the 
methoxyl at 6. Reaction of the bromoquinone with sodium azide, fol- 
lowed by reduction produced the desired 6-methoxy-7-aminoquinoline- 
5,8-dione. The difference in the displacements observed with ammonia 
versus azide was explained in terms of a cyclic intermediate. It was 
suggested that the azide anion initially attacked at the 6 position, 
followed by rearrangement to the 7-azido compound via a triazoline 
intermediate. 

Rao (35) (Figure 1.4, Scheme II), taking a different approach 
to the construction of the A-B system, used a modified Friedlander 
synthesis in the synthesis of tricyclic analogues of streptonigrin. 
Condensation of 3,5-dimethoxybenzaldehyde with 2-acetyl pyridine under 
basic conditions produced the chalcone. Nitration of the chalcone 
followed by reductive cyclization gave 6,8-dimethoxy-2(2'pyridyl )- 
quinoline. This compound was subjected to oxidative demethylation 
with nitric acid to produce the 6-hydroxy-7-nitroquinone which on 
reduction, followed by methylation with diazomethane gave the desired 
6-methoxy-7-amino-2(2'pyridyl)quinoline-5,8-dione. 

Hibino and Weinreb (40) (Figure 1.5, Scheme III) synthesized 
the same compound using the classical Friedlander route. Condensa- 
tion of 2-acetyl pyridine with 6-amino-2-benzenesulfonyloxy-3-methoxy- 
benzaldehyde under basic conditions produced 5-hydroxy-6-methoxy- 
2(2'pyridyl)quinoline. Oxidation of the hydroxy quinoline with 
Fremy's salt (potassium nitrosodisulfonate) gave the 6-methoxyquino- 
line-5,8-dione which was converted to the chloroquinone. Reaction 






11 



CH O 





u CH,0 

3 "OH v 3 





N0 2 S 2°4 v CH 3° 



» 




HNO 



-^ 




/-h7 ch.n. CH 3° 




Scheme II 
Figure 1.4. Synthesis of 2(2'-pyridyl )quinoline-5,8-dione. 



12 



CH 




^N 



^ 



~XH 3 JOH CH 3 



^^ 





CI, NO N- 



^> J -^ 



No 2 S 2 Q 4 C^ 




Scheme III 
Figure 1.5. Synthesis of 2(2'pyridyl )quinoline-5,8-dione. 



13 

with sodium azide followed by reduction gave the aminoquinone as 
before. 

The construction of the C-D ring portion of streptonigrin was 
more elusive. Kametani et al . (36) (Figure 1.6, Scheme IV) in an 
effort directed toward the synthesis of the C-D ring system, used a 
2-benzyloxy-3,4-dimethoxyphenyl-a,e-unsaturated ketone as a starting 
point. Reaction of this compound with ethyl cyanoacetate under 
Knoevenagel reaction conditions produced the pyridone which was 
converted to the chloropyridine with phenyl phosphonyl chloride. 

Liao et al. (42) (Figure 1.7, Scheme V), using a different 
starting material, a di ketone, produced the same chloropyridine as 
above. Further transformation of this compound by catalytic hydro- 
genation gave the pyridine. Reprotection of the phenol followed by 
oxidation of the 6-methyl group and protection of the resulting alde- 
hyde gave the cyclic acetal. Hydrolysis of the nitrile and Hoffman 
rearrangement gave the ami nopyri dine. Hydrolysis of the acetal fol- 
lowed by oxidation to the acid and deprotection of the phenol gave 
the C-D ring system of streptonigrin; however, this could not be used 
in the total synthesis of streptonigrin due to a lack of substitution 

at 2. 

The first total synthesis of streptonigrin was accomplished by 
Weinreb in Basha et al . (32) (Figure 1.8, Scheme VI) in 1980. The key 
feature of this work was the construction of the C-D rings system via 



14 




NH.OAc 



NK 



:ooEt 



OCH. 




Scheme IV 
Figure 1.6. Synthesis directed toward the C-D ring. 



15 






cu ifY H3 CJ ^-> [0j v H -^ H ^Ar^ 






TSOH mcJ^-^^ 




= B 



nqOH NaOCH. 

> — — *» 

2 H 




Scheme V 
Figure 1.7. Synthesis of the C-D ring of streptonigrin. 



16 




4 steps > ^° 

CHjO 



CH 3% 





Ba(OH) C^^^M^XOOH 
M> 




■+■ Isomer 



Xylene 




AcO 
COOH SOCI 2x MCPBA^ ACgO ^^^yCOOCH 



CH 3 OH 



L> > *-^ 



K 2 CQ 3 



» 



SOC1, 



^OH 



— * O ' 



t-BuOK 



COOH 
COOH 



thf/dmso THF/H O 




OOCH 3 CF,CO,H KMnO, 

3 3 > ±-» 




o" o* 

OOCH 3 (PhO) 2 PN 3 CH 3 n.^W\XoOCH 3 Ac2 q 

Scheme VI 



^> 2 3 > ?-> 



CH 3 OH 



Figure 1.8. First total synthesis of streptonigrin. 



17 




OOCH 3 L i-CH 2 -^(OCH 3 ) 2 ^ MnQ 2 <CH 3 0> 2 P 



hhsvCOOCH, 



+ 




NOjSjq, NaOCH 



h 




Fremy s 
Salt ^ 



IN, 



■* 



OOCH, 



non. Na 2 s 2 o 4 CH,a 
^ > 




AICU NHLOH 
- — £> — ^ Streptonigri n 



OCH- 



Scheme VI-- Continued 
Figure 1.8.— Continued. 



18 

a Di els-Alder adduct. After functionalization of the C-D system, 
using oxidations and a Sommelet-Hauser rearrangement, it was con- 
densed with 6-nitro-2-benzenesulfonyloxy-3-methoxybenzaldehyde to 
produce the nitro chalcone, which was subjected to reductive cycliza- 
tion. Oxidation of the phenol with Fremy's salt to produce the quin- 
one, followed by sequential treatment with IN3, sodium azide and 
sodium dithionite gave the aminoquinone. Deprotection of the phenol 
and carboxylic acid gave synthetic streptonigrin in 26 steps with an 
overall yield of 0.026%. 

In the second total synthesis published in 1981, Kende et al. 
(33) (Figure 1.9, Scheme VII) used a strategy similar to that of Liao 
et al. (42) in the construction of the C-D rings. The starting mate- 
rial again was a functional i zed phenyl e-di ketone which was converted 
to the e-keto enamine with ammonia. Condensation of this material 
with methyl acetoacetate gave the arylpyridone. Reduction of the 
acetyl group at 3, followed by reaction of the pyridone with phenyl 
phosphonyl chloride gave the 2-chloropyridine which was converted to 
the 2-nitrile. Reaction of the nitrile with methyl magnesium bromide 
gave the key intermediate, the 2-acetyl-3-vinyl-aryl pyridine. The 
construction of the tetracyclic skeleton of streptonigrin was accom- 
plished in the usual way. Oxidative cleavage of the 3-vinyl group 
followed by oxidation of the 6-methyl group gave the diacid which was 
selectively esterified at 6. Application of the Yamada modification 
of the Curtis rearrangement ((Ph0)2 PON3) gave the aminopyridine. 
The tetracyclic structure was converted in four steps to the qui none; 



^2^ V N^^'^3 




CH. 



CH 



CH- 




NaBH 4 PhPOCIj CuCN NCC-V|/T^y ch 3 CH 3 MgBr H 



-» 



^ > 



Rz CHjO^CHj 



-> ^ > 



:h 3 ^v n v :h 3 "'YV*"' t-euoK Rb v^y^i 



.-^H- 



TFA HN0 3 (CH3) 2 S0 4 

?> — ^ 7" 

K 2 C0 3 



CH 




1) OsO, 1) SeO, CHjOH 

2) NaI0 4 2)NaCI0 3 HCI 




N^ T COOCH 3 



Frea>y's _ u _ 
(PhO^N^ Na^q, ^ _salt > CH 3°V 



OOCH. 




Scheme VII 
Figure 1.9. Second total synthesis of streptonigrin. 



20 

however, the qui none was not converted to streptonigrin since this 
had been done previously by Weinreb in Basha et al . (32). The 
synthesis was carried out in 19 steps with an overall yield of 
approximately 1.3%. 

The last of the total syntheses of streptonigrin was performed 
by Boger and Panek (34) (Figure 1.10, Scheme VIII) in 1983. The key 
feature of this synthesis was the use of two consecutive inverse 
electron demand Di els-Alder reactions. The first, to construct the 
ABC rings: 1,2,4,5 tetrazine plus S-methyl thioimidate, followed by 
1,4,5 triazine plus morpholino enamine to construct the C-D system. 
The last compound in the scheme was the same as one synthesized by 
Kende et al. (33). Streptonigrin itself was not made. 

In studies conducted by Rao and Venkateswarlu (36) (Figure 
1.11, Scheme IX) to explore the synthesis and activity of tricyclic 
analogues of streptonigrin, the C-ring precursor was synthesized from 
2, 5-dimethyl pyridine. 

The key step in this synthesis was the production of the 2- 
acetyl pyridine from the ozonolysis of the benzylidene derivative fol- 
lowed by treatment of the resulting aldehyde with diazomethane. 



21 



ch 3 o 




"S^s TsSO.CI HNO, 



s) NoCN H 2 so 4 



-> 





°2 CH 3 



CH 



-> 




ch,o 2 c 



°2 CH 3 




PhSeNa CH 3 OH 



-> — = > 



HCI 



<PhO>,P-N, CH,I CH3 
- ^ _i > 



K 2 C0 3 




Scheme VIII 



Figure 1.10. Third total synthesis of streptonigrin. 



22 



CH,v^N\ 



NO 



NH, C^V^V^ DHONO ^CHO jS^V^V 3 " 



3> 



L ^^CH 3 ' 



» - > 

2) HNO- KOH 



2 N'^^A>l3 



P8r, CuCN 



■» » 



^^v^^-^cjn ch 1} 



j2TN0 2 j2fNO- 



0,N" 



"> 



3 > 



:H 3 H2 S °4 2J (CH^S 



A^ 



\ r C0 2 CH 3 CH N 



2 N' 



2 ,N 2 v C 






/K^K r co 2 CH 3 [/^^\co 2 c 



n>^^S:h 3 o 2 n>^^ch 3 



Scheme IX 
Figure 1.11. Synthesis of the C-ring precursor. 



CHAPTER II 
THE SYNTHESIS OF 1-PHENYL AND l-(2-PYRIDYL)-6-METH0XY-7- 
AMIM0IS0QUIN0LINE-5,8-DI0NE 

Structure-Activity Relationships 
In a series of chemical modifications carried out on strepto- 
nigrin (43) (Figure 2.1) (a) several interesting points came to 
light. The amino p-quinone of the A ring was found to be essential 
for activity; replacement of the amine by hydroxyl led to loss of 
activity as did reductive acetylation or methylation or conversion to 
an o-quinonoid system. In the B-ring, reduction to the tetrahydro 
derivative also caused loss of activity. Acetylation of the amino 
group of the C-ring led to an inactive compound but esterification of 
the carobxylic acid gave active analogues. In a series of tricyclic 
analogues it was found that the D-ring was not necessary for the 
antibacterial activity. 

In comparing the phenyl qui noline (Figure 2.1) (b)", the pyridyl- 
quinoline (Figure 2.1) (c) and the destrioxyphenyl streptonigrin 
(Figure 2.1) (g) in Figure 2.1, one would conclude that the pyridyl 
nitrogen was of less importance than the primary amine and the car- 
boxylic acid. The primary amine is much more basic than the pyridine 
nitrogen and thus can participate more readily in complexation with 
divalent metal ions (Lewis acids). It is also possible that the 
amine of the C-ring and the qui none carbonyl of the A-ring are 
necessary for binding to DNA through hydrogen bonding interactions. 

23 



24 



CH 



OOH 




a (RP = 1) 



CH 




b (RP = 0.25) 



CH 




c (RP = 0.25) 



^^ 



CH 





OOH 



%^ 




OOH 



CH., 



H 2 N^S^CH 3 



g (RP = 2) 




HzN e^ jQ 



Figure 2.1. Quinoline quinones, relative potency (RP) to 
streptonigrin. 



25 

In a study conducted by Lown et al . (24) the 2-(o-aminophenyl )- 
quinoline-5,8-dione (d) was comparable to streptonigrin in its 
ability to cause scission of DNA jji vitro . It was also found that 
the 2-(o-nitrophenyl ) precursor (h) showed a higher rate of cleavage 
in comparison to the 2-(o-aminophenyl ) compound. In vivo testing was 
not conducted on these compounds. 

In a recent study conducted by Boger et al. (44), to define the 
role of the carboxylic acid group in the activity of streptonigrin, 
the 6'-carboxy-2'-pyridyl- (e) and the 2'-amino-5'-carboxyphenylquin- 
oline-5,8-diones (f) were synthesized. The compounds with the car- 
boxyl were less active than those without the carboxyl against a num- 
ber of gram-positive and gram-negative bacteria and five tumor lines 
in cell culture. The methyl esters were more potent than the free 
carboxylic acids, although the carboxyl derivatives as a group were 
less potent than the unsubstituted compounds. The 2'aminophenyl com- 
pound (Figure 2.1) (d) was not tested. The carboxylic acid group 
would reduce the base strength of the amine in the carboxyaminophenyl 
derivative (Figure 2.1) (f); however, this does not explain the 
reduced potency of the pyridine carboxylic acid derivative (Figure 
2.1) (e) in comparison to the unsubstituted compound (Figure 2.1) 

(c). 

These two studies lend support to the importance of the role of 
the primary amino group in the mechanism of action of streptonigrin. 



26 

Other Heterocyclic and Carbocyclic Qui nones 
In a study conducted by Shaikh et al . (45) a series of hetero- 
cyclic (quinoline, isoquinoline, quinoxaline, quinazoline, and phtha- 
lazine) and naphthoquinones with electron withdrawing groups at the 6 
and 7 positions were synthesized, and their reduction potentials as 
well as their ability to cause DNA cleavage in vitro were determined. 
These workers found a good correlation between the reduction poten- 
tial of the qui nones and the extent of DNA cleavage. The order of 
activity was found to be as follows: naphthoquinone < quinoline 
quinone < isoquinoline quinone < quinoxaline quinone < quinazoline 
quinone < phthalazine quinone. In vivo testing of the compounds was 
not conducted. There does not seem to be a good correlation between 
DNA cleavage in vitro and in vivo activity as evidenced by the low 
activity shown by streptonigrin in the above study. Its potency in 
the DNA cleavage assay was found to be only slightly greater (0% vs. 
1%) than that of the naphthoquinones, whereas in cell culture, strep- 
tonigrin is one of the most potent compounds known. 

Several different types of heterocyclic fused p-benzoquinones 
(Figure 2.2) have also been synthesized (46-50) but none was compar- 
able to streptonigrin in cytotoxicity or antibacterial activity. One 
possible explanation for this low activity could be the absence of an 
ami noqui none system. 

In the modifications carried out on streptonigrin, very few 
changes could be made without loss of activity. One change that is 
likely to produce an active series of compounds is changing the 



27 



CHjf 




^NvN^ 4 ^ 




I I 



h 3 o-y^ 



M?\ 




^^^^S:h 






> V^V^ N ^>. 



I I 







Figure 2.2. Some fused heterocyclic qui nones. 



28 



B-ring from a quinoline to an isoquinoline. Isoquinoline and quinoline 
possess similar chemical and physical properties (51). The reduction 
potential of the isoquinoline-5,8-dione is slightly higher than that of 
the quinoline-5,8-dione (45) but as stated above, reduction potentials 
do not necessarily correlate with cytotoxicity. The isoquinoline ser- 
ies would lose the possibly favorable orientation present in the quino- 
line series, that of the proximity of the qui none carbonyl of the 
A-ring and the quinoline nitrogen of the B-ring (Figure 2.3). Thus, if 
this relationship is necessary for full activity, the isoquinolines 
should be less potent than the quinoline series of compounds, all other 
factors being equal. 

There were very few examples of isoquinoline-5,8-diones prior to 
this work being undertaken (Figure 2.4). As can be seen, they are 
either unsubstituted at the 1 position (52) (Figure 2.4 a, b) or sub- 
stituted at 1 but unsubstituted at 6 or 7 (53) (Figure 2.4 c, d). 
Also, these compounds were not tested for their biological activity. In 
addition to these synthetic compounds, a group of antitumor antibiot- 
ics, such as the saframycins (54), which are chemically fused tetrahy- 
droisoquinoline qui nones and a group of tetrahydroisoquinoline qui nones 
(55, 56) from the marine sponge Reneira sp . have been isolated. 

The compounds presented here are structurally similar to the sim- 
ple quinoline derivatives presented earlier. Only one component of the 
structure, the position of the nitrogen, has been changed. As the 
study progressed, only one change was made in the next compound synthe- 
sized so that the potency of the resulting compound could be related 
directly to the compound previous to it. This is the most meaningful 
way by which the relationship of structure to the activity can be 
determined. 



29 




CH 




dione. 



Figure 2.3. Comparison of quinolinedione to isoquinoline- 



30 







Figure 2.4. Some isoquinoline quinones synthesized 
previously. 



31 



Synthetic Approach to the Isoquinolines 
As stated previously, the unsubstituted phenyl and pyridyl 
derivatives of the quinoline system are of equal potency in the anti- 
bacterial assay. The first task at hand was to synthesize the pyri- 
dyl and phenyl derivatives of the isoquinoline (Figure 2.5) so that 
their potency could be compared to the quinoline series as well as to 
each other. 

Using the disconnection approach in Figure 2.6, the synthesis 
of the pyridyl isoquinoline was first visualized. The benzyl ether of 
o-vanillin was reacted with nitromethane and ammonium acetate in 
acetic acid at 50°C to form the nitrostyrene, which was reduced with 
lithium aluminum hydride in THF to give the phenyl ethyl amine. For 
the next step of forming the amide, difficulty in the formation of 
the acid chloride of picolinic acid required an alternative method 
(57), which used the hydrazide of picolinic acid formed via the ethyl 
ester. The hydrazide in combination with the phenyl ethylamine and 
triethylamine, when treated with NBS, gave the picolinarnide in good 
yield. The amide was treated with POCI3 in chloroform to give the 
dihydroisoquinoline although in poor yield with much decomposition. 
Oxidation of the dihydroisoquinoline with Pd/C in toluene gave the 
isoquinoline in quantitative yield. Cleavage of the benzyl ether was 
accomplished with HBr in acetic acid to give the phenol. Treatment 
of the phenol with Fremy's salt (potassium nitrosodi sulfonate) gave 
the qui none which was treated with bromine in chloroform to give the 
bromoquinone. Reaction of this material with sodium azide in DMF 
gave the azidoquinone, which was reduced with sodium dithionite in 
methanol /THF/H2O to give the desired ami noqui none. 



32 




CH 




Figure 2.5. l-Phenyl-6-methoxy-7-aminoisoquinoline-5,8-dione 
( 2.11 ) and l(2-pyridy1 )-6-methoxy-7-aminoisoquinoline-5,8-dione 
(2.17). 



33 




CH 





NH, 



OOH 



C 




Figure 2.6. Disconnection of l-(2-pyridyl )-6-methoxy-7-amino- 
i soqui nol i ne-5 ,8-di one. 






34 

Attempts to synthesize the 1-phenylisoquinoline derivative 
failed when the above reaction scheme was used, due to the inability 
of the dihydroisoquinoline to be oxidized to the isoquinoline with 
Pd/C. This problem and the low yields obtained in the formation of 
the 1-pyridyl dihydroisoquinoline required several modifications of 
the first scheme (Figure 2.7). 

1. The starting aldehyde was changed to 5-benzyloxy-2,3-di- 
methoxybenzaldehyde obtained from o-vanillin in 4 steps (58). The 
presence of electron donating groups in the aldehyde, both ortho and 
para to the position of attack, increased the yield from approximate- 
ly 10% to approximately 75%. 

2. The problem with oxidation of the 1-phenyl dihydroisoquino- 
line to the 1-phenylisoquinoline was overcome by introducing an 
easily eliminated group at the 4-position of the dihydroisoquinoline, 
as in a modified Pictet-Gam's isoquinoline synthesis (59). 

A methoxy group was introduced into the nitrostyrerie via a 
base-catalyzed addition to the double bond (60) to give a 2-methoxy- 
2-phenylnitroethane (2.3). Reduction of the nitro compound with 
lithium aluminum hydride gave the 2-phenyl -2-methoxyethyl amine 

Formation of the picolinamide and reaction with POCI3 in 
benzene gave the isoquinoline derivative as the major product, with a 
minor amount of the 4-methoxydihydroisoquinoline. In a similar fash- 
ion, the benzamide was synthesized via benzoyl chloride and reacted 
with POCI3 in benzene. However, in this case the 4-methoxydihy- 
droisoquinoline was the major product with approximately 30% as 



35 




2 NaOCH 



CH OH 



» CH3 °&^ >2 



2.1 



2.2 



ImI 



TMF 



CH. 



CH 35L^< H 3 

UAtH 4 y CH3 °K)l | BCq V 



A 






2^ 




poci. 



-> 



or CHC1. 



2.5 R=phenyl 
2.12 R=2-pyridyl 




2.6 R=phenyl 2/7 

R=2-pyridyl 2.1; 



HCI/CHOH CH O 
3 —> 




^ 



AcOH 



2.8 R=phenyl 
2.U R=2-pyridyl 





NoBH, 



thf/choh h 




2.9 R=phenyl 
2.15 R=2-pyridyl 



2.10 R=phenyl 
2.16 R=2-pyridyl 



2.11 R=phenyl 
2.17 R=2-pyridyl 



Figure 2.7. Synthesis of 1-phenyl- and l-(2-pyridyl )-6-methoxy- 
7-aininoisoquino1ine-5,8-dione. 



36 

the isoquinoline. This posed no real problem because during the 
cleavage of the benzyl ether with 6N-methanolic HC1 , the 4-methoxy 
group was also eliminated giving the 8-hydroxyisoquinoline as the 
only product. 

Oxidation of the 8-hydroxyisoquinoline with bromine in a mix- 
ture of chloroform-acetic acid gave the bromoquinone with the bromo- 
phenol as a minor component. The formation of the bromophenol could 
be eliminated by running the reaction in acetic acid alone. As 
before, reaction of the bromoquinone with sodium azide in DMF and 
reduction with sodium dithionite gave the aminoquinone. The reduc- 
tion of the azidoquinone could also be accomplished using sodium 
borohydride in THF/methanol . 

Experimental 

Nuclear magnetic resonance (NMR) spectra were recorded on a 
Varian EM 390 and chemical shifts ( 6 ) were reported in parts per mil- 
lion (pprn) relative to internal tetramethylsilane (0.0 ppm). Infra- 
red (IR) spectra were recorded on a Beckman Acculab 3 as KBr pellets 
and reported as cm"*. Melting points were determined on a Fisher- 
Johns hot stage melting point apparatus and are uncorrected. Ele- 
mental analysis was carried out by Atlantic Microlabs, Atlanta, GA. 
Alumina (Woelm) was used at activity grade III. 
2,3-Dimethoxy-5-nitrobenza1dehyde 

To a stirred ice-cold solution of o-vanillin (30 g) in 
chloroform (100 mL) was added 70% nitric acid (30 mL) dropwise. 
After all the nitric acid had been added, the reaction mixture was 
stirred for an additional 5 min. Ice was added and the chloroform 



37 

layer washed with water and 1 M phosphate buffer (pH=7). The chloro- 
form layer was dried and concentrated to give the nitrobenzaldehyde. 
A mixture of the phenolic nitrobenzaldehyde and potassium carbonate 
(25 g) in DMF (150 mL) was treated with dimethyl sulfate (30 mL) and 
the reaction mixture was stirred overnight. The reaction mixture was 
poured into ice water (1 L), precipitating the 2,3-dimethoxy-4-nitro- 
benzaldehyde which was filtered, 35 g (85%) m.p. 110-112°C. 
2,3-Dimethoxy-5-nitrobenzaldehyde dimethyl acetal 

A mixture of the benzaldehyde from above (30 g) and Dowex 
50H + (10 g) in 300 mL of methanol was refluxed for 24 h. The mix- 
ture was filtered and the methanol was concentrated to a small volume 
whereupon the acetal crystallized. The mixture was diluted with 
dilute NH4OH and the crystals filtered and dried, 35g (96%) m.p. 
91-93°C, Lit 90-92°C (58). 
2,3-Dimethoxy-5-hydroxybenza1dehyde 

A solution of the acetal (8 g) in THF (100 mL) and methanol (10 
mL) was subjected to catalytic hydrogenation in a Parr apparatus over 
5% Pd/C. The reaction was continued until no more hydrogen was taken 
up. The reaction mixture was filtered through celite to remove the 
5% Pd/C and the filtrate was concentrated to dryness. The residue 
was dissolved in cold 3N H2SO4 (100 mL) and the solution cooled 
to -10°C with an ice/methanol bath. To this stirred solution was 
slowly added, beneath the surface, a solution of 2.4 g (1.1 eq) of 
sodium nitrite in 20 mL of H2O. On completion of the addition, the 
excess nitrous acid was decomposed with ammonium sulfamate. 

This cold solution of the diazonium salt was added dropwise to 
a boiling solution of 5% sulfuric acid, 10% sodium sulfate (2 L). 



38 

Nitrogen evolution commenced immediately. The reaction mixture was 
cooled to room temperature and extracted twice with ethyl ether. The 
ethyl layer was washed with aqueous sodium bicarbonate, dried and 
concentrated to give the phenol 75% (4.25 g) m.p. 144-146°C, Lit 
143-145°C (58). 
5-Benzyloxy-2,3-dimethoxybenza1dehyde 2.1 

To a solution of 5-hydroxy-2,3-dimethoxybenzaldehyde (58) (10 
g) in DMF (50 mL) were added benzyl chloride (7.6 g 1.1 eq) and anhy- 
drous potassium carbonate (10 g 1.3 eq). The reaction mixture was 
stirred under reflux for 2 h, cooled and poured into water. The 
aqueous suspension was extracted with ethyl acetate, which was dried 
and concentrated. Crystallization from ligroin gave ZA as white 
needles, yield 13.4 g (90%); m.p. 55-56°C; NMR (Figure 2.8) (CDCI3) 
610.44 (1H, s, ArCHO), 7.43 (5H, bs, Ph), 7.00 (1H, d, J=5Hz, Ar), 

6.86 (1H, d, J=5Hz, Ar) , 5.07 (2H, s, C^Ph), 3.94 (3H, s, 0CH ), 

3.87 (3H, s, OCH3); IR (KBr) cm" 1 2880, 1680, 1600, 1480, 1380, 1340, 
1270, 1222, 1189, 1040, 985, 920, 840, 825, 740, 700, 690. Anal. 
Calc. for C 16 H 16 4 : C, 70.58; H, 5.8. Found: C, 70.6; H, 5.93. 
2,3-Dimethoxy-5-benzyloxy-g-nitrostyrene 2.2 

A mixture of 2.1 (10 g), nitromethane (10 mL) and ammonium 
acetate (10 g) in acetic acid (50 mL) was heated on a water bath 
(70°C) with stirring until TLC indicated the absence of starting 
material. The cooled reaction mixture was diluted with water and 
extracted with ethyl acetate. The solvent layer was washed twice 
with water and aqueous sodium bicarbonate, dried and concentrated. 
The residue was chromatographed over Florisil in benzene. Elution of 
the major band with benzene gave after crystallization from methanol, 



39 




L 



m 



is 




M 



3 
o 

E 
O 

u 



E 

a 
+-> 

cu 

Q- 



co 

CM 

<D 

S- 
3 



40 

the g-nitrostyrene as bright yellow needles, yield 10 g (87%); m.p. 
105-107°C; NMR (Figure 2.9) (CDCI3) 6 8.20 (1H, d, J=22.5Hz, 
H-C=C), 7.72 (1H, d, J=22.5Hz, C=C-H), 7.45 (5H, bs, Ph), 6.75 (1H, 
d, J=5Hz, Ar), 6.62 (1H, d, J=5Hz, Ar), 5.08 (2H, s, CH 2 Ph), 3.86 
(6H, s, 2xOCH 3 ); IR (KBr) cm" 1 2950, 1630, 1580, 1510, 1480, 1320, 
1285, 1180, 1040, 960, 820, 680. Anal. Calc. for C 17 H 17 5 N: C, 
64.76; H, 5.4; N, 4.4. Found: C, 64.83; H, 5.48; N, 4.39. 
2-(2,3-Dimethoxy-5-benzyloxyphenyl )-2-methoxy-l-nitroethane 2.3 

To a stirred suspension of 2^_2 (10 g) in methanol at -10°C, was 
slowly added freshly prepared 20% sodium methoxide solution (18 ml_). 
The reaction mixture was stirred until all the suspended material had 
gone into solution and TLC indicated the absence of starting mater- 
ial. The solution was acidified with acetic acid, diluted with water 
and extracted twice with ethyl acetate. The solvent layer was washed 
twice with aqueous sodium bicarbonate, dried and concentrated. 
Crystallization from methanol gave 2.3 as an off white sandy solid 
yield 10.5 g (95%); m.p. 95-96°C; NMR (Figure 2.10) (CDCI3) 6 7.43 
(5H, bs, Ph), 6.60 (2H, s, Ar), 5.28 (1H, t, J=10Hz, Ar, CH(0CH 3 ) 
CH 2 -N0 2 ), 5.05 (2H, s, CH 2 Ph), 4.59 (2H, d, J=10Hz, Ar-CH(0CH 3 ) 
CH2-N0 2 ), 3.84 (6H, s, 2xOCH 3 ), 3.30 (3H, s, 0CH 3 ); IR (KBr) (cm" 1 ) 
3005, 2980, 2970, 2920, 1590, 1550, 1490, 1460, 1380, 1370, 1240, 
1000, 860, 790, 680. Anal. Calc. for C 18 H 21 0gN: C, 62.24; H, 6.05; N, 
4.03. Found: C, 62.32; H, 6.11; N, 4.01. 
2-(2,3-Dimethoxy-5-benzyloxyphenyl )-2-methoxyethyl amine 2.4 

Into a 500 mL 2-neck round bottom flask fitted with an addition 
funnel and reflux condenser was placed lithium aluminum hydride (3 
g). Dry THF (50 mL) was added slowly and the suspension stirred for 
15 min. A solution of ^_3 (10 g) in THF (150 mL) was added dropwise; 
after all the starting material had been added, the addition funnel 



41 



J. 



fM 



J_ 



1 



IT 



s 



cvj 

^3 
C 

Q 

a. 

cs 
o 








3 

S_ 

+-> 
o 

Ol 

o. 
<n 

en 



CVJ 

O) 

s_ 



42 



-4- 




rx 



m 



-- (O 



CM 

C 
3 

O 



o 



s- 
+-> 
u 

0) 



CNJ 

0) 

a 



43 



was replaced with a stopper and the reaction mixture refluxed for 6-8 
h on a water bath. Water was cautiously added to the cooled reaction 
mixture to decompose the excess LAH. The resulting suspension was 
filtered through celite and the celite washed with THF. The THF 
was dried and concentrated to give the amine as an oil. The oil was 
converted to its oxalate salt with a saturated solution of oxalic 
acid in ethyl ether and filtered. The salt was recrystallized from 
MeOH/Acetone to give a white crystalline solid, yield 9.4 g (80%); 
m.p. 175-178°C slow; NMR (Figure 2.11) (DMS0d 6 ) 6 7.44 (5H, m, Ph), 
6.75 (1H, d, J=5Hz, Ar) , 6.53 (1H, d, J=5Hz, Ar), 5.65 (3H, g, NH 3 + ) , 
5.10 (2H, s, CHgPh), 4.77 (1H, t, J=llHz, ArCH(0CH 3 )CH 2 -NH 3 + ) , 3.81 
(3H, s, 0CH 3 ), 3.69 (3H, s, OCH3), 3.16 (3H, s, OCH3), 2.97 (2H, d, 
J=llHz, CH2NH 3 + ); IR (KBr) cm" 1 3450, 3005, 2980, 2950, 1710-1690, 
960. Anal. Calc. for C 18 H 23 4 N: C, 58.96; H, 6.14; N, 3.4. Found: 
C, 58.89; H, 6.22; N. 3.39. 
N-[2-(2,3-Dimethoxy-5-benzy1oxypheny1-2-methoxyethyl]benzamide 2.5 

A stirring suspension of ZA_ (2 g) and tri ethyl amine (1.5 g) in 
dichloromethane (50 mL) was treated dropwise with benzoyl chloride 
and the reaction followed by TLC. When TLC indicated the absence of 
the phenyl ethyl amine, water was added to decompose the excess acid 
chloride. The reaction mixture was poured into ethyl ether, washed 
twice with aqueous acid and aqueous sodium bicarbonate, dried and 
concentrated. The residue was chromatographed over silica gel in 
benzene and the amide was eluted with 2% acetone/benzene to give a 
viscous oil, yield 2 g (97%); NMR (CDCI3) 6 7.82 (2H, m), 7.43 
(8H, m), 6.92 (2H, s, Ar), 5.03 (2H, s, C^Ph), 4.85 (0.5H, d, 
J=4.5Hz), 4.77 (0.5H, d, J=4.5Hz), 3.85 (3H, s, OCH3), 3.83 (3H, s, 
OCH3), 3.6 (1H, d, J=4.5Hz), 3.4 (1H, d, J=4.5Hz), 3.27 (3H, s, 



44 



A- 



r-- 




• I 

c 
o 

Q. 



3 

u 



(XI 



en 



45 

0CH 3 ); IR (KBr) cm" 1 , 3080, 2950, 2850, 1650, 1600, 1490, 1460, 

1430, 1350, 1325, 1270, 1230, 1190, 1150, 1110, 1050, 1005, 830. 

Anal. Calc. for C 25 H 27 5 N: C, 71.25; H, 6.4; N, 3. Found: C, 

71.20; H, 6.46; N, 3.28. 

1-Phenyl -5,6-dimethoxy-8-benzy1 oxyi soqui nol i ne 2.6 and 
l-Pheny1-4,5,6-trimethoxy-8-benzy1oxy-3,4-dihydroisoquino1ine 2.7 

To a solution of 2.5 (4 g) in benzene (50 mL) was added phos- 
phorus oxychloride (POCI3) (15 mL) and the reaction mixture was 
stirred under reflux for 4 h. TLC indicated the absence of the amide 
and the presence of two basic compounds. The reaction mixture was 
cooled to room temperature, poured over ice, to decompose the excess 
POCI3 and was stirred until all the ice had melted. The acidic 
aqueous solution was washed twice with ethyl ether, made basic with 
cold 20% sodium hydroxide and extracted twice with chloroform. The 
chloroform layer was dried and concentrated to give the crude mixture 
of 2^6 and 2J_> 3 g (82%). A portion of this mixture was placed on 
an alumina column in benzene. Elution with benzene gave the i soqui n- 
oline 2.6 . Further elution with 2% acetone in benzene gave the 
4-methoxydihydroi soqui nol ine 2.7 . The ratio of i soqui nol ine to 
4-methoxydihydroi soqui nol ine was 1:3. 

2.6.— A white crystalline solid; m.p. 117-119°C; NMR (CDCI3) 
6 8.51 (1H, d, J=6Hz, C-4), 7.86 (1H, d, J-6Hz, C-3), 7.43 (3H, m, 
Ph), 7.23 (5H, m, Ar), 6.86 (2H, m, Ph), (1H, s, C-7), 4.84 (2H, s, 
CH 2 Ph), 3.95 (3H, s, OCH3), 3.93 (3H, s, OCH3); IR (KBr) cm" 1 
3080, 2960, 2870, 1610, 1550, 1440, 1390, 1360, 1340, 1235, 1190, 
1125, 1050, 995, 830, 815, 790, 760, 730, 695. 

2. 7 . —Converted to the hydrochloride salt with ethanolic HC1 , 
recrystallized from acetone as a greenish-yellow crystalline solid; 
m.p. 195-200°C (slow loss of HC1 ) ; NMR (DMSOdg) (poorly resolved) 6 



46 

7.53 (5H, m), 7.17 (4H, m), 6.73 (1H, s), 6.63 (1H, s), 5.05 (2H, s), 
4.83 (1H, s), 4.40 (1H, s), 4.23 (1H, s), 4.07 (3H, s), 3.83 (3H, s), 
3.30 (3H, s); IR (KBr) cm -1 3000-2600, 1610, 1585, 1495, 1455, 1425, 
1340, 1320, 1270, 1245, 1220, 1135, 1080, 1040, 980, 810, 740, 690. 
1-Phenyl -5 ,6-dimethoxy-8-hydroxy i sogui nol i ne 2.8 

The crude mixture of 2^6 and 1J_ (2 g) was treated with 6N 
methanolic hydrochloric acid (50 mL) and stirred under reflux over- 
night; TLC showed no starting compounds. The reaction mixture was 
cooled, diluted with water, washed with ethyl ether and neutralized 
with sodium bicarbonate. The neutralized aqueous layer was extracted 
with chloroform which was dried and concentrated to give 2J3, crys- 
tallized from ethyl ether/1 i groin as an orange-yellow solid, yield, 
1.22 g (85%); m.p. 225-227°C; NMR (CDCI3) 6 8.35 (1H, d, J=6Hz, 
C-4), 7.34 (1H, d, J=6Hz, C-3), 7.38 (5H, m, Ph), 6.77 (1H, s, C-7), 
3.90 (3H, s, OCH3), 3.83 (3H, s, OCH3). IR (KBr) cm -1 3010, 2950, 
1610, 1560, 1475, 1450, 1400, 1360, 1265, 1225, 1170, 1175, 1100, 
1030, 990, 845, 815, 770, 700, 665. Anal. Calc. for C 17 H 15 3 N: C, 
72.59; H, 5.34; N, 4.98. Found: C, 71.10; H, 5.33; N, 4.82. 
1-Phenyl -6-methoxy-7-bromoi sogui nol i ne-5 ,8-di one 2.9 

Excess bromine (1 mL) was added to a solution of 2^8 (0.32 g) in 
acetic acid and the reaction mixture was stirred overnight. When TLC 
indicated the absence of starting material, the reaction mixture was 
poured into chloroform and an ice cold solution of sodium bisulfite 
was added to decompose the excess bromine. The layers were separated 
and the chloroform layer was washed twice with aqueous sodium bicar- 
bonate, dried and concentrated. The residue crystallized from ethyl 
ether to give 2^9. as a bright yellow solid, yield 0.357 g (91%) m.p.; 



47 

160°C dec; NMR (CDCI3) 6 9.04 (1H, d, J=6Hz, C-4), 7.97 (1H, d, 
J=6Hz, C-3), 7.47 (5H, s, Ph), 4.31 (3H, s, OCH3); IR (KBr) cm" 1 
3080, 7980, 1670, 1650, 1585, 1555, 1435, 1390, 1330, 1245, 1100, 
1030, 910, 835, 750, 740, 690. Anal. Calc. for C 16 H 10 3 N Br: 
C, 55.97; H, 2.91; N, 4.08. Found: C, 55.88; H, 2.91; N, 4.01. 
l-Pheny1-6-methoxy-7-azidoisoguinoline-5,8-dione 2.10 

To a stirred solution of 2^9 (0.2 g) in DMF (10 mL) was added 
sodium azide (0.041 g, 1.1 eq). After 15 min TLC showed completion 
of reaction, the reaction mixture was diluted with water and extract- 
ed twice with ethyl acetate. The ethyl acetate layer was washed 
three times with water, dried and concentrated. The residue crystal- 
lized from ethyl ether as orange-yellow crystalline solid, yield 
0.156 g (88%); m.p. 130°C dec; NMR (CDCI3) 6 9.O (1H, d, J=6Hz), 
7.97 (1H, d, J=6Hz), 7.45 (5H, s, Ph), 4.15 (3H, s); IR (KBr) cm" 1 
3080, 2965, 2120, 1660, 1585, 1555, 1435, 1390, 1330, 1245, 1100. 
1-Phenyl -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 2.11 

A solution of 2.10 (0.1 g) in THF/methanol (10 mL/0.5 mL) was 
treated with solid sodium borohydride until TLC indicated the absence 
of the azidoquinone. The reaction mixture was diluted with water to 
decompose the excess borohydride and extracted twice with ethyl 
acetate. The ethyl acetate layer was washed with water, dried and 
concentrated. The residue was placed on an alumina column in 
chloroform and the aminoquinone was eluted with the same solvent. 
The product 2.11 crystallized from ethyl ether/1 i groin as a bright 
red crystalline solid; yield 0.081 g (90%); m.p. 168-170°C; NMR 
(Figure 2.12) (DMS0d 6 ) 6 8.96 (1H, d, J=6Hz), 7.89 (1H, d, J=6Hz), 



48 




C\J 

■o 
e 

o 
a. 



i. 
+-> 
u 

s. 






o> 



49 

7.0 (2H, b), 3.82 (3H, s); IR (KBr) cm" 1 3480, 3280, 3140, 2950, 
1680, 1635, 1610, 1550, 1360, 1290, 1230, 1100. Anal. Calc. for 
C 16 H 12 3 N 2 : C, 68.57; H, 4.28; N, 10. Found: C, 68.27; H, 4.34; 

N, 9.91. 

N-[2-(2,3-dimethoxy-5-benzyloxyphenyl)-2-methoxyethyl]picolin- 
amide 2.12 

Method A .— To a stirred suspension of 2^4 (2 g), tri ethyl amine 
(2.5 g, 5 eq) and picolinoyl hydrazide in dichloromethane (50 mL) at 
-10°C was added N-bromosuccinimide (1.8 g, 2 eq) in small portions. 
The reaction mixture was stirred for an additional 15 min, poured in- 
to ethyl ether and washed with aqueous acid and aqueous sodium bicarb- 
onate. The ether layer was dried, concentrated and the residue was 
placed on a Florisil column in benzene. The amide was eluted with 2% 
acetone in benzene to give a viscous oil, yield, 1.6 g (80%). 

Method B .--A suspension of picolinic acid (5 g) in thionyl 
chloride (10 mL) was heated in an oil bath held at 80°C until all the 
picolinic acid had gone into solution and then for an additional 
hour. The thionyl chloride was removed under reduced pressure and 
the acid chloride was used without further purification. 

A stirred suspension of 2A_ (2 g) and triethylamine (1.5 g, 3 
eq) in dichloromethane (50 mL) was treated with picolinoyl chloride 
dropwise until TLC indicated the absence of the phenyl ethyl amine. 
Water was added to decompose the excess acid chloride and the reac- 
tion mixture poured into ethyl ether. The ether layer was washed 
twice with aqueous acid and aqueous sodium bicarbonate, dried and 
concentrated. The residue was chromatographed over Florisil in 
benzene and the amide eluted with 2% acetone in benzene to give a 
viscous oil, yield, 1.9 g (92%); NMR (CDC1 3 ) 6 8.57 (1H, dd, J=7.5 



50 

+ 1.5Hz, Py), 7.85 (1H, td, J=7.5 + 1.5Hz, Py), 7.43 (5H, Ph + 2H, 
Py, m), 6.64 (2H, s, Ar), 5.17 (1H, dd, J=7.5 + 2.5Hz), 5.13 (2H, s, 
CH^Ph), 4.07 (1H, dd, J=7.5 + 2.5Hz), 4.0 (1H, dd, J=7.5 + 2.5Hz), 
3.83 (6H, s, 2xOCH 3 ), 3.33 (3H, 0CH 3 ); IR (KBr) cm' 1 2940, 2840, 
1675, 1595, 1520, 1490, 1460, 1440, 1330, 1220, 1190, 1150, 1105, 
1040, 1000, 750, 690. Anal. Calc. for C 24 H 26 5 N 2 : C, 68.25; H, 6.16; 
N, 6.63. Found: C, 68.11; H, 6.21; N, 6.60. 
l-(2-Pyridy1 )-5,6-dimethoxy-8-benzy1oxyisoquino1 ine 2.13 

To a solution of the picolinamide, 2.12 (1.6 g) in chloroform 
(50 mL) was added phosphorus oxychloride (POCI3, 10 mL) and the 
solution was stirred under reflux for 5-6 hrs. The absence of start- 
ing material and the presence of two basic compounds were indicated 
by TLC. The cooled reaction mixture was poured over ice and stirred 
until all the ice had melted, and the excess P0C1 3 decomposed. 
This acidic aqueous solution was washed twice with ethyl ether, made 
basic with cold 20% sodium hydroxide and extracted twice with chloro- 
form. The chloroform was dried and concentrated to give 1.2 g (88%) 
of a crude mixture of the isoquinoline 2.13 and the 4-methoxydihy- 
droisoquinoline, which was not characterized. Trituration with ethyl 
ether gave 2.13 as a tan crystalline solid, yield 0.9, g (66%); m.p. 
130-132°C; NMR (CDCI3) <5 8.48 (1H, d, J=6Hz, C-4), 8.41 (1H, dd, 
J=4.5 + 1.5Hz), 7.90 (1H, d, J=6Hz, C-3), 7.43 (2H, m), 7.25 (3H, m), 
6.96 (4H, m), 6.70 (1H, s, C-7), 4.86 (2H, s, CH^Ph), 3.93 (3H, s, 
OCH3), 3.90 (3H, s, OCH3); IR (KBr) cm' 1 3080, 3040, 3009, 2950, 
2850, 1610, 1580, 1550, 1450, 1440, 1390, 1370, 1360, 1340, 1265, 
1240, 1198, 1120, 1110, 1070, 1050, 1040, 990, 960, 930, 910, 885, 
850, 835, 805, 790, 690, 660. 



51 

1- ( 2-Pyri dyl ) -5 , 6-dimethoxy-8-hydroxyi soqui nol i ne 2.14 

Conversion of 2.13 (0.6 g) to 2.14 was carried out as described 
under 2.8 . The product was obtained from ethyl ether as an orange 
crystalline solid, yield 0.352 g (77%), ra.p. 102-103°C; NMR (DMSOdg) 
6 8.65 (1H, dd, J=7.5 + 1.5), 8.43 (1H, d, J=6Hz C-4), 803 (1H, td, 
J=7.5 + 1.5Hz), 7.91 (1H, d, J=6Hz), 7.52 (2H m), 6.86 (1H, s, C-7), 

3.97 (3H, s, 0CH 3 ), 3.86 (3H, s, OCH3); IR (KBr) cm -1 3080, 3020, 
2980, 2950, 2850, 1620, 1590, 1550, 1480, 1440, 1410, 1390, 1370, 
1320, 1280, 1235, 1235, 1200, 1130, 1110, 1055, 1010, 980, 835, 810, 
795, 770, 740, 675, 650. Anal. Calc. for C 16 H 14 3 N 2 : C, 68.08; 

H, 4.96; N, 9.92. Found: C, 67.99; H, 5.02; N, 9.88. 
l-(2-Pyridy1 ) -6-methoxy-7-bromoi soqui nol ine-5,8-di one 2.15 

Conversion of 2.14 (0.3 g) to 2.15 was carried out as described 
under 2.9 . The product was crystallized from ethyl ether to give 
bright orange-yellow needles, yield 0.32 g (90%); m.p. 150-160°C dec; 
NMR (CDCI3) 6 9.05 (1H, d, J=6Hz, C-4), 8.66 (1H, dd, J=7.5 + 
15Hz), 8.02 (1H, d, J=6Hz, C-3), 7.90 (1H, td, J=7.5 + 1.5Hz), 7.47 
(2H, m), 4.32 (3H, s, OCH3); IR (KBr) cm' 1 3080, 2980, 1680, 1655, 
1580, 1560, 1440, 1400, 1325, 1245, 1120, 1070, 1040, 1020, 990, 920, 
840, 790, 760, 655. 
l-(2-Pyridyl )-6-methoxy-7-azidoi soqui no! ine-5,8-dione 2.16 

Conversion of 2.15 (0.2 g) to 2.16 was carried out as described 
under 2.10 . The product separated from ethyl ether/1 i groin as an 
orange-red crystalline solid, yield 0.153 g (86%); m.p. 140°C dec; 
NMR (CDCI3) 6 9.0 (1H, d, J=6Hz), 8.63 (1H, dd, J=7.5 + 1.5Hz), 

7.98 (1H, d, J=6Hz), 7.90 (1H, td, J=7.5 + 1.5Hz), 7.53 (2H, m), 4.18 
(3H, s, OCH3); IR (KBr) cm-1 3080, 2965, 2120, 1655, 1585, 1560, 



52 

1480, 1445, 1410, 1320, 1320, 1285, 1250, 1180, 1140, 1130, 1090, 

1050, 960, 920, 750. 

l-(2-Pyridyl )-6-methoxy-7-aminoisoquinol ine-5,8-dione 2.17 

Conversion of 2.16 (0.12 g) to 2.17 was carried out as 
described under 2.11 . The product was crystallized from ethyl ether 
to give a blue-black solid, yield 0.09 g (82%); m.p. 160-162°C; NMR 
(Figure 2.12) (DMS0d 6 ) « 8.99 (1H, d, J=6Hz), 8.56 (1H, dd, J=7.5 + 
1.5Hz), 7.95 (1H, d, J=6Hz), 7.93 (1H, td, J=7.5 + 1.5Hz), 7.52 (2H, 
m), 7.03 (2H, bs, NH 2 ) , 3.83 (3H, s, 0CH 3 ); IR (KBr) cm" 1 
3440, 3320, 2960, 1675, 1630, 1590, 1545, 1440, 1410, 1355, 1240, 
1000, 790. Anal. Calc. for C^H^Nj: C, 64.05; H, 3.9; N, 14.95. 
Found: C, 63,89; H, 3.99; N, 14.90. 






53 




— 
e 

o 

Q. 

5 

CJ 



■1) 



CO 



CM 
CD 

s_ 

3 



CHAPTER III 
SYNTHESIS OF NITRO AND AMINOPHENYLISOQUINOLINE DERIVATIVES 

Introduction 

In comparing the phenyl qui noline and the pyridyl qui noline 
derivatives to streptonigrin (Chapter II, Fig. 2.1), one would 
conclude that the pyridyl nitrogen is of less importance than the 
primary amine and the caroboxylic acid groups. The primary amine is 
much more basic than the pyridine nitrogen and thus can participate 
more readily in complexations with divalent metal ions (Lewis acids). 
It is also possible that the amine and the qui none carbonyl can 
participate in binding to DNA through hydrogen-bonding interactions. 

It was of interest to determine how the substitution of an 
amine at the ortho, meta, and para positions of the phenyl ring in 
the isoqui noline system would affect the potency in comparison to the 
phenyl itself. All three of the positional isomers are necessary to 
determine the influence of the positional differences between iso- 
qui noline and qui noline, specifically, the intramolecular distances 
between the amine and the quinone carbonyl. The relationship of the 
amine and the aromatic nitrogen should be the same for the quinoline 
and isoquinoline, although in the case of the quinoline, this site is 
possibly influenced by the quinone carbonyl (Figure 3.1). 

If metal binding between the primary amine and the aromatic 
nitrogen is necessary for activity, this could take place in the 
o-amino derivative of the phenyl isoquinoline. However, if activity 
is expressed, not through binding of metal but through a hydrogen 

54 



55 




CH 




Figure 3.1. Favorable spatial arrangement of a quinoline 
qui none compared to isoquinoline qui none. 






56 

bonding interaction with DNA, the meta and para amino derivatives 
would be in a better position to participate (Figure 3.2). 

This substitution could also influence the electronics of the 
system and possibly the reduction potential of the qui none. The 
amine, an electron donating group, would add electron density to the 
heterocyclic quinone system thus making the quinone more difficult to 
reduce but permitting auto-oxidation to take place more readily. The 
nitro precursors to the amine derivatives would have the opposite 
effect; electron withdrawing groups would decrease the electron dens- 
ity of the quinone thus making the quinone more easily reduced and 
contributing less to auto-oxidation. 

The synthesis (Figure 3.3) of the amino derivatives of the 
phenyl isoquinoline begins with ortho, meta or para nitrobenzoyl 
chloride prepared from the acid with thionyl chloride. Reaction of 
the acid chloride with the phenyl ethyl amine ( 2.4 ) and triethylamine 
in dichloromethane gave the amide in good yield. Treatment of the 
amide in benzene or, as in the case of the meta-nitro amide where 
solubility was a problem, in chloroform with phosphorus oxychloride 
gave a mixture of isoquinoline and 4-methoxydihydroisoquinoline. The 
para nitro compound gave only the isoquinoline. Cleavage of the 
benzyl ether and elimination of the 4-methoxy group were accomplished 
with 6N-methanolic hydrochloric acid to give the 8-hydroxyisoqui no- 
line which was oxidized to the bromoquinone with bromine in acetic 
acid. Reaction of the bromoquinone in DMF with sodium azide gave the 
azidoquinone. The azidoquinone was reduced with sodium borohydride 



57 





Figure 3.2. Possible favorable spatial arrangement of the 
isoquinolines. 






58 



CH 




CH. 



KQ 



CH 



hH 2 (EtljN/CHjClj 



CH 




POCI, 

- — M> 

& or CHCI 



3.1 R=p-nitrophenyl 

3.9 R=m-nitrophenyl 

3.18 R=o-nitrophenyl 




R=p-nitrophenyl 3.2 
3-10 R=m-nitrophenyl 3.11 
3.19 R=o-nitrophenyl 3.20 



6N 
HCI/CHOH CH O 
1 i> 3 




Br 



2 ■> 



AcOH 



3.3 R=p-nitrophenyl 
3.12 R=m-nitrophenyl 
3.21 R=o-nitrophenyl 




NoN 




NoBH. 



3.4 R=p-nitrophenyl 
3-13 R=m-nitrophenyl 
3.22 R=o-nitrophenyl 



THF/CH OH 



3.5 R=p-nitrophenyl 
3. H R=m-nitrophenyl 
3.23 R=o-nitrophenyl 




3.6 R=p-nitrophenyl 
3.15 R=m-nitrophenyl 
3.25 R=o-aminophenyl 




3.24 R=o-nitrophenyl 



<- 



j "*£' 



Na 2 S 2 4 



2)H 2 o/AcOH 



CM 




3.7 R=p-aminophenyl 
3.16 R=m-amniophenyl 



Figure 3.3. Synthesis of amino- and nitrophenyl compounds. 



59 



in THF/methanol to give the nitrophenyl ami noqui none in the case of 
the meta and para nitro compounds but gave the o-aminophenyl 
ami noqui none with the o-nitro compound. 

When this reaction was followed more carefully during a second 
run, it was found that initially the borohydride reduced the azide to 
the amine, with a small amount of the o-aminophenyl derivative being 
produced. On addition of more borohydride the reaction went to com- 
pletion to produce the o-aminophenyl aminoquinone. One possible 
explanation for this reaction is the close proximity of the quinone 
carbonyl to the nitro group; coordination of the borohydride anion 
between the carbonyl and the nitro group may facilitate the nitro 
group being reduced (Figure 3.5). In the case of the meta and para 
nitro compounds this is not possible due to the greater distance 
between the carbonyl and the nitro group. 

The o-nitrophenyl aminoquinone was synthesized using triphenyl- 
phosphine (61) to form the tri phenyl phosphinimine which was then 
hydrolyzed with water/acetic acid to give the aminoquinone. 

The meta and para amino phenyl were synthesized by reaction of 
the nitrophenyl azidoquinone with sodium dithionite in THF/water/- 
methanol. 

The structures of the ami noqui nones synthesized are presented 
in Figure 3.4. 

Experimental 

Nuclear magnetic resonance (NMR) spectra were recorded on a 
Varian EM 390 and chemical shifts (6) were reported in parts per mil- 
lion (ppm) relative to internal tetramethylsilane (0.0 ppm). Infra- 
red (IR) spectra were recorded on a Beckman Acculab 3 as KBr pellets 
and reported as cm" . Melting points were determined on a Fisher- 
Johns hot stage melting point apparatus and are uncorrected. 



60 



CH 





Figure 3.4. Structure of m'tro- and aminophenylisoquinoline 
ami noqui nones. 



61 




Figure 3.5. Reduction of nitro group by borohydride. 



62 

Elemental analysis was carried out by Atlantic Microlabs, Atlanta, 

GA. Alumina (Woelm) was used as activity grade III. 

N-[2-(2,3-Dimethoxy-5-benzyloxypheny1)-2-methoxyethyl] 
(4 nitro)benzamide 3.1 

To a stirred suspension of 2A_ (2 g) and triethylamine (1.5 g 
3 eq) in dichloromethane (50 mL) was added 4-nitrobenzoyl chloride 
dropwise. When TLC indicated the absence of the phenyl ethyl amine, 
water was added and the mixture stirred for 30 min to decompose the 
excess acid chloride. The reaction mixture was diluted with ethyl 
ether, and the ether layer washed twice with aqueous acid and aqueous 
sodium bicarbonate, dried and concentrated to give a pale yellow 
crystalline solid from ethyl ether/1 i groin 3.1 , yield 2.6 g (90%); m.p. 
133-134°C; NMR (CDC1 3 ) & 8.30 (2H, d, J=8.4Hz), 7.94 (2H, d, 
J=8.4Hz), 7.41 (5H, bs, Ph), 6.59 (2H, s, Ar), 5.04 (2H,s, O^Ph), 
4.84 (0.5H, d, J=4.5Hz), 4.76 (0.5H, d, J=4.5Hz), 3.85 (6H, s, 
2xOCH 3 ), 3.64 (1H, d, J=4.5Hz), 3.52 (1H, d, J=4.5Hz), 3.28 (3H, s, 
OCH3); IR (KBr) cm" 1 3340, 3120, 3090, 2995, 2960, 1640, 1590, 1520, 
1480, 1425, 1340, 1320, 1250, 1225, 1185, 1135, 1105, 1060, 1040, 
1000, 950, 940, 870, 830, 735, 690. Anal. Calc. for C 25 H 26 7 N 2 -H 2 0: 
C, 62.24; H, 5.81; N, 5.81. Found: C, 62.54; H, 5.50; N, 5.77. 
l-( 4 ' Ni trophenyl ) -5 ,6-dimethoxy-8-benzyl oxyi soqui no! i ne 3.2 

To a solution of 3.1 (1 g) in benzene (50 mL) was added phos- 
phorus oxychloride (10 mL) and the solution stirred under reflux 
overnight. When TLC indicated the absence of the starting amide, the 
reaction mixture was cooled and poured over ice to decompose the 
excess P0C1 3. This acidic aqueous solution was washed twice with 
ethyl ether, made basic with cold 20% sodium hydroxide and extracted 
with chloroform. The chloroform layer was dried and concentrated to 



63 



give a yellow crystalline solid from ethyl ether, 3.2 yield 0.644 g 
(72%), m.p. 164-166°C; NMR (CDCI3) 6 8.49 (1H, d, J=6Hz) , 7.95 (1H, 
d, J=6Hz), 7.90 (2H, d, J=8.7Hz), 7.43 (2H, d, J=8.7Hz), 7.22 (3H, m), 
6.92 (2H, m), 6.79 (1H, s), 4.80 (2H, s), 4.05 (3H, s), 3.97 (3H, s); 
IR (KBr) cm" 1 3000, 2950, 2860, 1610, 1555, 1520, 1450, 1400, 1350, 
1235, 1130, 1110, 1040, 990, 935, 850, 830, 810, 760, 695. 
l-(4 ' Ni trophenyl ) -5 ,6-dimethoxy-8-hydroxyi soqui nol i ne 3.3 

A solution of 3^2 (1 g) in 6N methanolic hydrochloric acid (50 
mL) was refluxed until TLC indicated the absence of starting mater- 
ial. The reaction mixture was cooled, diluted with water and washed 
twice with ethyl ether to remove the benzyl chloride. The acidic 
aqueous layer was neutralized with sodium bicarbonate and extracted 
with chloroform, which was dried and concentrated to give a dull 
orange-brown crystalline solid from ethyl ether 3.3 , yield 0.65 g 
(83%), m.p. 205°C dec; NMR (CDCI3) 6 8.44 (1H, d, J=6Hz), 8.28 (2H, 
d, J=8.7Hz), 7.83 (1H, d, J=6Hz), 7.71 (2H, d, J=8.7Hz), 6.86 (1H, s), 
3.94 (3H, s), 3.84 (3H, s); IR (KBr) cm" 1 3480, 2980, 1610, 1520, 1400, 
1350, 1240, 1220, 1200, 1130, 1110, 1050, 990, 840, 805, 750. Anal. 
Calc. for C 17 H 14 5 N 2 : C, 62.58; H, 4.3; N, 8.59. Found: C, 61.43; 
H, 4.31; N, 8.24. 
l-(4 ' Nitrophenyl ) -6-methoxy-7-bromoi soqui nol i ne-5 ,8-di one 3.4 

Excess bromine (1 mL) was added to a solution of 3^3 (0.34 g) 
in acetic acid (15 mL) and the solution stirred overnight. The reac- 
tion mixture was diluted with chloroform and the excess bromine 
decomposed with an ice-cold solution of sodium bisulfite. The chlor- 
oform layer was washed twice with aqueous sodium bicarbonate, dried 
and concentrated to give a bright yellow crystalline solid from ethyl 
ether 3^4, yield 0.318 g (85%); m.p. 195°C, dec; NMR (DMSOdg) 6 9.08 



64 

(1H, d, J=6Hz), 8.30 (2H, d, J=8.7Hz), 8.00 (1H, d, J=6Hz), 7.69 (2H, 

d, J=8.7Hz), 4.25 (3H, s) ; IR (KBr) cm" 1 3095, 2960, 1665, 1580, 

1550, 1500, 1435, 1340, 1250, 1100, 1020, 920, 850, 830, 765, 750, 735, 

680. Anal. Calc. for C 16 H 9 5 N 2 Br: C, 49.38; H, 2.33; N, 7.20. 

Found: C, 49.38; H, 2.37; N, 7.17. 

1- ( 4 ' Ni trophenyl ) -6-methoxy-7-azi doi sogui nol i ne-5 ,8-di one 3.5 

To a solution of 3^4 (0.255 g) in DMF (20 mL) was added sodium 
azide (0.047 g, 1.1 eq) and the reaction mixture stirred for 30 min 
at which time TLC indicated the reaction was complete. The reaction 
mixture was diluted with water and extracted twice with ethyl ace- 
tate. The ethyl acetate layer was washed three times with water, 
dried and concentrated to give an orange- brown crystalline solid from 
ethyl ether 3^, yield 0.195 g (85%); m.p. 150°C, dec; NMR (DMS0d 6 ) 
6 9.12 (1H, d, J=6Hz), 8.33 (2H, d, J=8.7Hz), 8.02 (1H, d, J=6Hz), 
7.84 (2H, d, J=8.7Hz), 4.09 (3H, s); IR (KBr) cm" 1 3090, 2120, 
1660, 1600, 1560, 1510, 1440, 1350, 1280, 1240, 1100, 855, 740. 
l-( 4 ' Ni trophenyl ) -6-methoxy-7-ami noi sogui nol i ne-5 ,8-di one 3^6 

A solution of 3^5 (0.1 g) in THF (20 mL) and methanol (0.5 mL) 
was treated with solid sodium borohydride in small portions until TLC 
indicated the absence of starting material. The reaction mixture was 
diluted with water to decompose the excess borohydride and extracted 
with ethyl acetate, which was dried and concentrated. The residue was 
chromatographed over alumina in chloroform and the major band was elut- 
ed with chloroform to give a reddish-purple crystalline solid from eth- 
yl ether, 3^, yield 0.083 g (90%) m.p. 270-271°C; NMR (Figure 3.6) 
(DMSOdg) 6 9.02 (1H, b), 8.31 (2H, d, J=8.7Hz), 7.95 (1H, d, J=6Hz) , 



65 




i 



3 



^ 



• I 

ro| 

o 

D. 

E 
o 
o 

4- 

o 



v£ 




E 

t_ 
+-> 
u 

<D 
la- 
in 

as 



to 

CO 



an •- 

CD 



0> 



66 

7.79 (2H, d, J=8.7Hz), 7.04 (2H, bs), 3.83 (3H, s); IR (KBr) cm" 1 , 
3480, 3365, 3090, 2968, 1675, 1620, 1560, 1510, 1440, 1410, 1350, 1240, 
1100, 1000, 920, 860, 800, 740, 690. Anal. Calc. for C 16 H n 5 N 3 : 
C, 59.07; H, 3.38; N, 12.92. Found: C, 59.01; H, 3.40; N, 12.87. 
l-(4 ' Ami nophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.7 

To a solution of 3^ (0.2 g) in THF/water/methanol (20 mL, 
1:1:0.1) was added sodium dithionite (2 g) and the reaction mixture 
stirred under reflux for 30 min. When TLC indicated the absence of 
starting material, 2N sulfuric acid (50 mL) was added and the reaction 
mixture was refluxed for an additional 40 min. The cooled reaction 
mixture was filtered to remove precipitated sulfur, washed once with 
ether and made slightly basic. The basic aqueous layer was extracted 
twice with ethyl acetate, which was dried and concentrated. The resi- 
due was placed on an alumina column in chloroform and the major band 
eluted with chloroform to give a purple-red crystalline solid from eth- 
yl ether, 3^7., yield 0.084 g (50%), m.p. 253-255°C; NMR (Figure 3.7) 
(DMS0d 6 ) 6 8.85 (1H, d, J=6Hz), 7.72 (1H, d, J=6Hz), 7.30 (2H, d, 6.97 
(2H, b), 6.61 (2H, d, J=8.7Hz), 5.41 (2H, b), 3.81 (3H, s); IR (KBr) 
cm" 1 3460, 3365, 3180, 1660, 1630, 1610, 1560, 1555, 1440, 1395, 
1360, 1295, 1230, 1175, 1100, 990, 975, 835, 820, 760, 745. Anal. 
Calc. for C 16 H 13 3 N 3 -l/3 H 2 0: C, 63.8; H, 4.51; N, 13.95. Found: 
C, 64.06; H, 4.47; N, 13.95. 
l-( 4 ' Acetyl ami nophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.8 

Acetic anhydride (5 mL) was added to 3^ (0.06 g) and the 
reaction mixture warmed to complete solution. After standing over 
night, water was added to decompose the acetic anhydride and the 
reaction mixture was extracted with ethyl acetate. The ethyl acetate 
layer was washed twice with aqueous sodium bicarbonate, dried and 



67 



_J_ 




r\j 



n 



1/1 



IS 



CO 

-a 

o 

a. 
E 
o 
o 



E 

Z3 
S- 

-t-J 
u 
(1) 
a. 



0) 
i- 

en 



00 



a> 



68 



concentrated. The residue was chromatographed over alumina in chlor- 
oform and the major band eluted with chloroform to give a red crys- 
talline solid from ether, 3^8, yield 0.064 g (93%) m.p. 286-288°C; 
NMR (Figure 3.8) (DMS0d 6 ) 6 8.94 (1H, d, J=6Hz), 7.85 (1H, d, 
J=6Hz), 7.77 (2H, d, J=9Hz), 7.47 (2H, d, J=9Hz), 7.00 (2H, b), 3.84 
(3H, s), 2.12 (3H, s); IR (KBr) cm" 1 3480, 3310, 3280, 1665, 1635, 
1610, 1540, 1440, 1410, 1390, 1270, 1230, 1100, 990, 925, 835, 815, 
760, 740. Anal. Calc. for C 18 H 15 4 N 3 -l/2 H 2 0: C, 62.42; H, 4.62; 
N, 12.13. Found: C, 62.34; H, 4.42; N, 11.98. 

N-[2-( 2,3-Dimethoxy-5-benzyl oxyphenyl ) -2-methoxyethyl ]3 ' nitro- 
benzamide 3.9 

Conversion of 2.4 (2.5 g) to 3.9 using 3-nitrobenzoyl chloride 

was carried out as under 3.1 . The product was crystallized from 

ether as a pale yellow crystalline solid, 3.9 , yield 2.5 g (88%), 

m.p. 90-92°C; NMR (CDC1 3 ) 6 8.62 (1H, m), 8.23 (2H, m) 7.65 (1H, 

m), 7.43 (5H, m), 6.61 (2H, s), 5.06 (2H, s), 4.87 (0.5H, d, 

J=4.5Hz), 4.77 (0.5H, d, J=4.5Hz), 3.88 (3H, s), 3.87 (3H, s), 3.67 

(1H, d, J=4.5Hz), 3.58 (1H, d, J=4.5Hz), 3.30 (3H, s); IR (KBr) 

cm" 1 3440, 3090, 2940, 2840, 1660, 1600, 1530, 1495, 1470, 1430, 1350, 

1270, 1220, 1195, 1180, 1165, 1110, 1060, 840, 740, 710. Anal. Calc. 

for C 25 H 26 7 N 2 : C, 64.38; H, 5.58; N, 6.00. Found: C, 64.44; H, 5.61; 

N, 5.95. 

l-( 3 ' Ni trophenyl ) -5 ,6-dimethoxy-8-benzyl oxyi soqui no! i ne 3.10 

and l-( 3 'ni trophenyl 1 )-4,5,6-trimethoxy-8-benzyloxy-3,4-dihydro- 

i soqui noline 3.11 

To a solution of 3.9 in chloroform (50 mL) was added phosphorus 

oxychloride (15 mL) and the reaction mixture stirred under reflux 

overnight. When TLC indicated the absence of starting material and 

the presence of two basic compounds, the cooled reaction mixture was 



69 




CO 



\n 



-a 
c 

Z3 

o 

a. 




— * 



0< 



■M 

O) 
Q. 
to 



00 
CO 

aj 



70 

poured over ice to decompose the excess POCI3. The acidic aqueous 
solution was washed twice with ethyl ether, made basic with cold 20% 
sodium hydroxide and extracted with chloroform. The chloroform 
layer was dried and concentrated to give a mixture of 3.10 and 3.11 , 
yield 1.9 g (99%), which was used directly in the next step. A por- 
tion of this mixture was chromatographed over alumina in benzene to 
obtain analytical samples; 3.10 was eluted with benzene and 3.11 was 
eluted with 2% acetone/benzene. 

3A0. —Orange-yellow crystals from ethyl ether; m.p. 124-126°C, 
NMR (CDCI3) 6 8.53 (1H, d, J=6Hz), 3.20 (1H, m), 7.96 (1H, d, 
J=6Hz), 7.77 (2H, m), 7.28 (4H, m), 6.93 (2H, m), 6.82 (1H, s), 4.83 
(2H, s), 4.06 (3H, s), 3.99 (3H, s); IR (KBr) cm" 1 3030, 2960, 
2870, 1605, 1550, 1515, 1440, 1385, 1340, 1240, 1115, 1090, 1050, 
1000, 880, 810, 735, 710, 680. 

3.11 . — A bright yellow HC1 salt from ethyl ether, m.p. 195- 
199°C (loss of HC1); NMR (DMSOdg) 6 8.34 (1H, m), 8.03 (2H, m), 7.62 
(1H, m), 7.19 (4H, m), 6.8 (2H, m), 5.01 (2H, s), 4.89 (1H, bs), 4.45 
(1H, bs), 4.27 (1H, bs), 4.12 (3H, s), 3.87 (3H s), 3.35 (3H, s); IR 
(KBr) cm" 1 2950, 2840, 2780, 1615, 1585, 1520, 1480, 1340, 1320, 
1290, 1260, 1230, 1080, 1035, 980, 690. 
l-(3'Nitrophenyl )-5,6-dimethoxy-8-hydroxyisoquino1ine 3.12 

Conversion of the mixture of 3.10 and 3.11 (1.5 g) to 3.12 was 
carried out as under 3.3 . The product, a yellow-orange crystalline 
solid was recrystallized from ether, yield 0.7 g (60%); m.p. 215°C, 



71 



dec; NMR (CDCl 3 /DMSOdg) 6 8.37 (1H, d, J=6Hz), 8.27 (1H, m), 8.16 (1H, 
m), 7.87 (1H, m), 7.83 (1H, d, J=6Hz), 7.58 (1H, m), 6.82 (1H, s), 3.94 
(3H, s), 3.87 (3H, s); IR (KBr) cm" 1 3090, 2950, 1610, 1560, 1530, 1485, 
1455, 1400, 1350, 1235, 1110, 1050, 1000, 950, 940, 830, 810, 740, 710, 690. 
Anal. Calc. for C 17 H 14 5 N 2 : C, 62.58; H, 4.3; N, 8.59. Found: C, 62.51; 
H, 4.3; N, 8.52. 
l-( 3 ' Ni trophenyl ) -6-methoxy-7-bromoi soqui nol i ne-5 ,8-di one 3.13 

Conversion of 3.12 (0.303 g) to 3.13 was carried out as under 3.4 . 
The product, a bright yellow crystalline solid, was recrystallized from 
ethyl ether, yield 0.321 g (89%); m.p. 198-200°C; NMR (DMSOdg) 6 9.14 
(1H, d, J=6Hz), 8.34 (2H, m), 8.05 (1H, d, J=6Hz), 7.83 (2H, m) , 4.26 
(3H, s, 0CH 3 ); IR (KBr) cm' 1 3120, 2980, 1670, 1580, 1525, 1350, 1340, 
1280, 1120, 1075, 1025, 930, 860, 830, 800, 765, 730, 670. Anal. Calc. 
for C 16 H 9 5 N 2 Br: C, 49.48; H, 2.3; N, 7.22. Found: C, 49.45; H, 2.35; 
N, 7.13. 
l-( 3 'Ni trophenyl ) -6-methoxy-7-azidoi soqui noli ne-5, 8-di one 3.14 

Conversion of 3.13 (0.31 g) to 3.14 was carried out as under 3^5 
using 0.057 mg of sodium azide. The product, a red-orange crystalline 
solid, was recrystallized from ethyl ether, yield 0.26 g (93%); m.p. 
130°C dec; NMR (DMSOdg) 6 9.10 (1H, d, J-6Hz), 8.4 (2H, m), 8.06 (1H, 
m), 8.00 (1H, d, J=6Hz), 7.78 (1H, m), 4.08 (3H, s); IR (KBr) cm" 1 3100, 
2960, 2120, 1675, 1595, 1550, 1525, 1435, 1340, 1280, 1240, 1110, 1070, 
1040, 720. 
1- ( 3 ' Ni trophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.15 

Conversion of 3.14 (0.1 g) to 3.15 was carried out as under 3.4 . 
The product, a red crystalline solid was recrystallized from ethyl 
ether, yield 0.083 g (90%); m.p. 264-266°C; NMR (Figure 3.9) (DMSOdg) 



72 



(M 



n 




iT 



Lf) 



CO 



c 



o 

CL 



o 
u 



\S 



o 

Q. 
in 



00 



en 



0) 



Oi 






73 



6 9.03 (1H, d, J=6Hz), 8.43 (2H, m), 7.97 (1H, d, J=6Hz), 7.71 (2H, 
m), 7.11 (2H, b), 3.83 (3H, s); IR (KBr) cm" 1 3500, 3380, 1675, 1595, 
1550, 1520, 1435, 1400, 1340. Anal. Calc. for C 16 H n 5 N 3 : C, 59.07; 
H, 3.38; N, 12.92. Found: C, 59.23; H, 3.41; N, 12.95. 
l-( 3 ' Ami nophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.16 

The procedure was the same as that described under 3.7 ; from 
0.22 g of 3.14 , yield 0.085 g (46%). The product, 3.16 , a red 
crystalline solid, was recrystall ized from ethyl ether; m.p. 215- 
217°C; NMR (Figure 3.10) (DMSOdg) 6 8.47 (1H, d, J=6Hz), 7.79 (1H, d, 
J=6Hz), 7.0 (4H, m), 6.93 (2H, m), 5.03 (2H, b), 3.78 (3H, s); IR 
(KBr) cm' 1 3480, 3360, 3200, 2960, 1675, 1630, 1595, 1550, 1445, 
1400, 1360, 1310, 1240, 1195, 1090, 1000, 930, 790, 769, 740, 690. 
Anal. Calc. for C 16 H 13 3 N 3 : C, 65.08; H, 4.41; N, 14.23. Found: 
C, 65.13; H, 4.46; N, 14.22. 
1- ( 3 ' Acetyl ami nophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.17 

Acetic anhydride (2 mL) was added to 3.16 (0.046 g) and the reac- 
tion mixture was heated on a water bath for 30 min. Water was added to 
decompose the acetic anhydride and the reaction mixture extracted with 
ethyl acetate. The ethyl acetate layer was washed twice with aqueous 
sodium bicarbonate, dried and concentrated. The residue was chromato- 
graphed over alumina in chloroform and the major band eluted with 
chloroform to give a red crystalline solid from ether m.p. 263-264°C; 
NMR (Figure 3.11) (DMSOdg) 6 8.97 (1H, d, J=6Hz), 7.90 (1H, d, J=6Hz), 
7.67 (2H, m), 7.37 (1H, m), 7.18 (1H, m), 7.03 (2H, b), 3.84 (3H, s), 
2.07 (3H, s); IR (KBr) cm -1 3440, 3390, 3320, 1680, 1635, 1595, 1550, 
1435, 1405, 1360, 1230, 1000. Anal. Calc. for C 18 H 15 4 N 3 'l/6 H 2 0: 
C, 63.5; H, 4.5; N, 12.35. Found: C, 63.4; H, 4.52; N, 12.32. 



74 




75 




'a? 



CO 



o 

CL 



O 

o 



1 



X! 



3 

4-> 

o 



*\ 



ro 

s_ 

CD 



76 



N-[2-(2,3-Dimethoxy-5-benzyloxyphenyl)-2-methoxyethyl](2'nitro)- 
benzamide 3.18 

The procedure was the same as that described under 3.1 using 3 g 

of 2.4 . The product was purified by column chromatography on Florisil 

in benzene. The amine was eluted with 2% acetone in benzene to give a 

viscous oil 3.18 , yield 3.26 g (95%); NMR (CDC1 3 ) 68.03 (1H, n), 7.66- 

7.30 (8H, m), 6.93 (2H, s), 5.05 (2H, s), 4.87 (0.5H, d, J=4.5Hz), 4.77 

(0.5H, d, J=4.5Hz), 3.83 (6H, s), 3.54 (1H, d, J=4.5Hz), 3.46 (1H, d, 

J=4.5Hz), 3.26 (3H, s); IR (KBr) cm -1 3330, 3100, 2960, 2860, 1660, 

1600, 1530, 1490, 1450, 1350, 1320, 3220, 1190, 1170, 1150, 1110, 1050, 

1000, 850, 830, 780, 750, 690. Anal. Calc. for C 25 H 26 7 N 2 -3/4 H 2 0: 

C, 62.55; H, 5.73; N, 5.83. Found: C, 62.77; H, 5.62; N, 5.98. 

l-(2'Nitrophenyl)-5,6-dimethox,y-8-benzy1oxyisoquinoline 3.19 and 

l-( 2 ' Ni trophenyl ) -4 ,5 ,6-trimethoxy-8-benzyl oxy-3 ,4-di hydroi sogui no! i ne 

3.20 

A procedure was used similar to that described under 3^2, from 2 
g of 3.18 , yield 1.86 g (96%) of a mixture of 3.19 and 3.20 was ob- 
tained, which was used in the next step without further purification. 
A portion of the mixture was chromatographed over alumina in benzene to 
obtain analytical samples. The i sogui noline 3.19 was eluted with ben- 
zene and further elution with 2% acetone/benzene gave 3.20 . 

3.19 .— A bright yellow crystalline solid (ethyl ether) m.p. 188- 
190°C; NMR (CDCI3) 6 8.49 (1H, d, J=6Hz), 7.93 (1H, d, J=6Hz), 7.66 
(1H, m), 7.30 (5H, m), 7.00 (3H, m), 6.70 (1H, s), 4.75 (2H, s), 3.99 
(3H, s), 3.97 (3H, s); IR (KBr) cm" 1 2960, 2860, 1605, 1550, 1510, 
1450, 1350, 1230, 1110, 990, 930, 830, 810, 750, 690. 

3. 20 . —A light yellow crystalline solid (ether); m.p. 197°C, dec; 
NMR (CDCI3) 6 7.57 (2H, m) , 7.33 (4H, m), 6.87 (3H, m), 6.60 (1H, s), 
4.90 (1H), 4.67 (2H, s), 4.33 (1H), 4.00 (3H, s), 3.93 (3H, s), 3.46 
(3H, s); IR (KBr) cm" 1 2960, 2850, 1610, 1580, 1510, 1480, 1420, 1380, 
1340, 1320, 1290, 1260, 1240, 1130, 1080, 1040, 970, 890, 840, 740, 
690. 



77 

l-( 2 ' Ni trophenyl ) -5 ,6-dimethoxy-8-hydroxyi soqui nol i ne 3.21 

Conversion of 3.19 and 3.20 (1.5 g) to 3.21 was carried out as 
under 3.3 . The product, an orange-yellow crystalline solid was 
crystallized from ethyl ether, yield 0.655 g (60%); m.p. 165-167°C; 
NMR (DMS0d 6 ) 6 8.35 (1H, d, J=6Hz) , 8.13 (1H, dd, J=7.5 + 1.5Hz), 
7.81 (1H, d, J=6Hz), 7.65 (2H, td, J=7.5 + 1.5Hz), 7.43 (1H, dd, J=7.5 
+ 1.5Hz), 6.71 (1H, s), 3.90 (3H, s), 3.87 (3H, s); IR (KBr) cm -1 2950, 
1605, 1550, 1510, 1460, 1440, 1380, 1340, 1230, 1120, 1100, 1035, 980, 
825, 800, 750, 690. Anal. Calc. for C 17 H 14 5 N 2 : C, 62.58; H, 4.3; 
N, 8.59. Found: C, 62.44; H, 4.36; N, 8.53. 
l-( 2 ' Ni trophenyl ) -6-methoxy-7-bromoi soqui nol i ne-5 ,8-di one 3.22 

The procedure used was the same as that described under 3.4 , 
from 0.315 g of 3.21 , yield 0.308 g (82%) as a greenish-yellow 
crystalline solid (ethyl ether); m.p. 172-174°C; NMR (CDC1 3 ) 6 9.05 
(1H, d, J=6Hz), 8.3 (1H, dd, J=7.5 + 1.5Hz), 8.03 (1H, d, J=6Hz), 
7.72 (2H, m), 7.33 (1H, dd, J=7.5 + 1.5Hz), 4.33 (3H, s); IR (KBr) 
cm" 1 3140, 2960, 2860, 1675, 1595, 1560, 1520, 1440, 1400, 1345, 
1320, 1240, 1110, 1070, 1020, 920, 850, 830, 780, 750, 710, 690. 
l-( 2' Ni trophenyl )-6-methoxy-7-azidoisoquino1 ine-5,8-dione 3.23 

Conversion of 3.22 (0.2 g) to 3.23 was carried as described 
under JL5 using 0.039 g (1.1 eq) of sodium azide. The product, 3.23 , 
an orange-red crystalline solid was crystallized from ethyl ether, 
yield 0.15 g (83%); m.p. 115°C dec; NMR (CDC1 3 ) 6 9.03 (1H, d, 
J=6Hz), 8.30 (1H, m), 8.03 (1H, d, J=6Hz), 7.73 (2H, m), 7.36 (1H, 
m), 4.23 (3H, s); IR (KBr) cm" 1 3140, 3100, 2980, 2880, 2120, 
1660, 1600, 1550, 1440, 1395, 1350, 1330, 1240, 1070, 940, 730. 



78 

l-( 2 ' Ni trophenyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 3.24 

To a stirred solution of 3.23 (0.1 g) in dichloromethane (20 mL 
was added triphenylphosphine (0.075 g 1.1 eq). The solution immedi- 
ately changed from orange-yellow to deep purple and the evolution of 
nitrogen was observed. After 30 min, TLC indicated the absence of 
starting material, the reaction mixture was concentrated, redissolved 
in chloroform and chromatographed over alumina to separate the tri- 
phenylphosphine from the tri phenyl phosphinimine formed. 

The tri phenyl phosphinimine was dissolved in THF (1 mL) and 10% 
acetic acid (10 mL) was added to hydrolyze the imine. The reaction 
mixture was stirred for 1 h, diluted with water and extracted with 
ethyl acetate. The ethyl acetate layer was washed with aqueous 
sodium bicarbonate, dried and concentrated. The residue (containing 
the aminoquinone and triphenylphosphine oxide) was placed on an 
alumina column in chloroform and 3.24 was eluted with chloroform. 
The product was crystallized from ethyl ether as a dark purple-red 
solid, yield 0.055 g (60%); m.p. 198-200°C; NMR (Figure 3.12) 
(DMS0d 6 ) 6 8.99 (1H, d, J=6Hz), 8.77 (1H, m), 7.97 (1H, d, J=6Hz), 
7.81 (2H, m), 7.50 (1H, m), 6.96 (2H, b), 3.85 (3H, s); IR (KBr) 
cm" 1 3500, 3390, 3120, 2950, 2860, 1660, 1620, 1585, 1540, 1510, 
1430, 1400, 1350, 1330, 1295, 1230," 1050, 980, 910. Anal. Calc. for 
C 16 H n 5 N 3 : C, 59.07; H, 3.38; N, 12.92. Found: C, 59.23; H, 3.54; 
N, 12.84. 
l-( 2 ' Ami nophenyl ) -6-methoxy-7-ami noi soqui no! i ne-5 ,8-di one 3.25 

A solution of 3.23 (0.1 g) in THF (20 mL) and methanol (0.5 mL) 
was treated with solid sodium borohydride until TLC indicated the 
presence of only one compound. The reaction mixture was diluted with 



"V" 



79 



1 




OJ 



; .. 



m 



■a 

c 

3 
O 



= 

3 
E_ 

+-> 

CD 

c 
1/1 



ro 

0) 
S- 



80 

water to decompose the excess borohydride and extracted with ethyl 
acetate. The ethyl acetate layer was washed twice with water, dried 
and concentrated. The residue was chromatographed over alumina in 
chloroform and the major band eluted with chloroform to give a red 
crystalline solid from ethyl ether, 3.25 yield 0.076 g (90%); m.p. 
sub. 220, 250°C; NMR (Figure 3.13) (DMS0d 6 ) 6 9.34 (1H, d, J=6Hz), 
9.02 (1H, m), 8.91 (1H, m), 8.77 (1H, d, J=6Hz), 7.97 (2H, m), 7.14 
(2H, b), 3.92 (3H, s); IR (KBr) cm" 1 3480, 3360, 1630, 1610, 
1540, 1440, 1220, 760, 720. Anal. Calc. for C 16 H 12 2 N 3 : C, 68.8; 
H, 4.3, N; 15.05. Found: C, 69.29; H, 4.0; N, 15.12. 



81 




(M 



in 



is 



LO 



en 



3 
o 

o. 



=3 
+J 

u 

& 



00 



V 

i. 



C* 



CHAPTER IV 
SYNTHESIS OF HOMO ANALOGUES: 1-BENZYL, l-(2'NITRO)BENZYL, AND 
l-(2'AMIN0)BENZYL-6-METH0XY-7-AMIN0IS0QUIN0LINE-5,8-DI0NE 

Introduction 



The synthesis (Figure 4.1) of benzyl isoquino1ine-5,8-dione was of 
interest in order to determine the effect of moving the phenyl ring out 
of conjugation with the heterocyclic quinone system. The phenyl ring 
should act as an electron donating group, thus increasing the electron 
density of the heterocyclic system, and making the quinone less easily 
reduced but undergo auto-oxidation more readily. The benzyl group should 
also act as an electron donor albeit weaker, but it would lack the conju- 
gation ability of the phenyl ring. The benzyl isoquinoline system is 
present in many natural products with biological activity (Figure 4.2). 

The next modification was the introduction of an ortho-nitro or 
amino group on the aromatic ring. The substituents should have yery 
little, if any, effect on the heterocyclic quinone ring but may offer a 
site for metal chelation or biological interaction of other kinds. 

Experimental 

Nuclear magnetic resonance (NMR) spectra were recorded on a Var- 
ian EM 390 and chemical shifts (6) were reported in parts per million 
(ppm) relative to internal tetramethylsilane (0.0 ppm). Infrared (IR) 
spectra were recorded on a Beckman Acculab 3 as KBr pellets and reported 
as cm" 1 . Melting points were determined on a Fisher-Johns hot stage 
melting point apparatus and are uncorrected. Elemental analysis was car- 
ried out by Atlantic Microlabs, Atlanta, GA. Alumina (Woelm) was used at 
activity grade III. 

82 






33 



CH,Q OCH. 



CH 




»ca 



CH 







POC1 3 
& or CHCI 



2A 



4.1 R=benzyl 

L.l R=2-nitrobenzyl 




R=benzyl £±2 
4.8 R=2-nitrobezyl £^ 



6 H 



CH 



HCI/CH OH CH 
— > 3 




Br. 



AcOH 



4.3 R=benzyl 
4.10 R=2-nitrobenzyl 



CH 3 



9r 




NoN. 



DMF 




^ 2 1 



4.4 R=benzyl 
4.1 1 R=2-nitrobenzyl 



4- . 5 R=benzyl 
4.12 R=2-nitrobenzyl 



4.6 R=benzyl 
4.13 R=2-nitrobenzyl 



N02S 2 4 



CH 



^> 




4.14 R=2-aminobenzyl 



Figure 4.1. Synthesis of benzyl derivatives. 



84 




Figure 4.2. Structure of benzyl, 2-nitrobenzyl and 2-amino- 
benzyl isoquinoline ami noqui nones. 






85 

N-[2-(2,3-Dimethoxy-5-benzy1oxy-phenyl )-2-methoxyethyl]- 
phenylacetamide 4.1 

To a stirred suspension of 2.4 (2 g) and tri ethyl amine (1.5 g, 
3 eq) in dichloromethane (50 mL) was added phenylacetyl chloride 
dropwise until TLC indicated the absence of the phenylethyl amine. 
Water was added to decompose the excess acid chloride and the reac- 
tion mixture stirred for 30 min. After dilution with ethyl ether, 
the ether layer was washed twice with aqueous acid and aqueous sodium 
bicarbonate, dried and concentrated. The residue was chromatographed 
over Florisil in benzene and the amide eluted with 2% acetone in ben- 
zene to give 4U as a viscous oil, yield 2 g (94%); NMR (CDCI3) 
6 7.42 (5H, m), 7.31 (5H, m), 6.56 (1H, d, J=3Hz), 6.47 (1H, d, J=3Hz), 
5.01 (2H, s), 4.66 (0.5H, d, J=4.5Hz), 4.57 (0.5H, d, J=4.5Hz), 3.25 
(1H, d, J=4.5Hz), 3.17 (3H, s); IR (KBr) cm" 1 2940, 2840, 1640, 
1580, 1480, 1445, 1420, 1315, 1210, 1180, 1160, 1140, 1100, 1010, 990. 
Anal. Calc. for C 2 6H 2 90 5 N: C, 71.52; H, 6.67; N, 3.2. Found: C, 
71.56; H, 6.76; N, 3.19. 
1-Benzyl -5 ,6-dimethoxy-8-benzy1 oxyi soqui no! i ne 4.2 

A solution of 4^_1 (1.5 g) in benzene (50 mL) was treated with 
phosphorus oxychloride (10 mL) and the reaction mixture stirred under 
reflux overnight, when TLC indicated the absence of amide and the 
presence of two basic compounds. The cooled reaction mixture was 
poured over ice to decompose the excess POCI3 and stirred until all 
the ice had melted. The reaction mixture was washed twice with ethyl 
ether, made basic with cold 20% sodium hydroxide and extracted with 
chloroform which was dried and concentrated to give a mixture of 4.2 
and the 4-methoxydihydroi soqui no! ine (1.1 g). This mixture was 






86 

used in the next step without purification. An analytical sample of 
4.2 was obtained from preparative TLC on alumina eluted with 10% 
acetone in benzene. The 4-methoxydihydroisoquinoline was not 
characterized. 

4.2.--A tan crystalline solid (ethyl ether/1 i groi n) ; m.p. 
126-128°C; NMR (CDC1 3 ) 6 8.43 (1H, d, J=6Hz), 7.80 (1H, d, J=6Hz), 
7.25 (8H, m), 6.93 (2H, m) , 6.64 (1H, s), 5.10 (2H, s), 4.85 (2H, s), 
3.87 (5H, s); IR (KBr) cm" 1 3080, 3040, 2950, 1610, 1560, 1450, 
1350, 1240, 1210, 1190, 1120, 1100, 1060, 1010, 960, 840, 800, 740, 
720, 700. 
l-Benzyl-5,6-dimethoxy-8-hydroxyisoquinol ine 4.3 

A solution of the mixture from jLj? (1 g) in 6N methanol ic 
hydrochloric acid (50 mL) was stirred under reflux until TLC indicat- 
ed the presence of one compound. The cooled reaction mixture was 
diluted with water, washed twice with ethyl ether and neutralized 
with sodium bicarbonate. The neutralized aqueous layer was extracted 
with chloroform which was dried and concentrated. The residue was 
placed on an alumina column in chloroform and the major band eluted 
with chloroform to give a tan crystalline solid from ethyl ether, 
4.3 , yield 0.526 g (71%); m.p. 224-225°C; NMR (CDCl 3 /DMS0d 6 ) 
6 8.25 (1H, d, J=6Hz), 7.65 (1H, d, J=6Hz), 7.20 (5H, bs), 6.84 (1H, 
s), 4.85 (2H, s), 3.90 (3H, s), 3.78 (3H, s); IR (KBr) cm" 1 3000, 
1610, 1560, 1480, 1450, 1350, 1250, 1220, 1195, 1060, 1010, 820, 736, 
720. Anal. Calc. for C 18 H 17 3 N: C, 73.2; H, 5.76; N, 4.75. Found: 
C, 73.06; H, 5.82; N, 4.72. 






87 

1-Benzyl -6-methoxy-7-bromoi soqui nol i ne-5 ,8-di one 4.4 

Excess bromine (1 mL) was added to a solution of 4^2 (0.3 g) in 
acetic acid (15 mL) and the reaction mixture was stirred overnight. 
The reaction mixture was diluted with chloroform and an ice-cold 
solution of aqueous sodium bisulfite was added to decompose the 
excess bromine. The chloroform layer was washed with aqueous sodium 
bicarbonate, dried and concentrated to give a greenish-yellow crystal- 
line solid from ethyl ether, 4.4 , yield 0.308 (85%); m.p. 85-88°C; 
NMR (CDCI3) 6 8.98 (1H, d, J=6Hz), 7.85 (1H, d, J=6Hz), 7.28 (5H, 
m), 4.81 (2H, s), 4.28 (3H, s); IR (KBr) cm" 1 3020, 2960, 1650, 
1565, 1430, 1320, 1225, 1035, 770, 730, 700, 690. Anal. Calc. for 
C 17 H 12 3 N Br: C, 57.14; H, 3.36; N, 3.92. Found: C, 57.05; H, 3.36; 
N, 3.90. 
l-Benzy1-6-methoxy-7-azidoisoquinoline-5,8-dione 4.5 

To a solution of 4_A (0.2 g) in DMF (20 mL) was added sodium 
azide (0.040 g 1.1 eq)and the reaction mixture was stirred until TLC 
indicated the absence of starting material. The reaction mixture was 
diluted with water and extracted with ethyl acetate, which was washed 
three times with water, dried and concentrated. The product, an 
orange-red crystalline solid, was recrystallized from ethyl ether 4.5 
yield 0.161 g (90%) m.p. 84-86°C; NMR (CDCI3) 6 8.97 (1H, d, 
J=6Hz), 7.86 (1H, d, J=6Hz), 7.29 (5H, m), 4.79 (2H, s), 4.18 (2H, 
s); IR (KBr) cm" 1 2120, 1750, 1585, 1550, 1315, 1225, 1055, 690. 
1-Benzyl -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 4.6 

A stirred solution of 4.5 (0.1 g) in THF (20 mL) and methanol 
(0.5 mL) was treated with solid sodium borohydride until TLC indi- 
cated the absence of starting material. Water was added to decompose 



88 



the excess borohydride and the reaction mixture extracted with ethyl 

acetate. The ethyl acetate layer was washed with water, dried and 

concentrated to give a bright red crystalline solid from ethyl ether, 

4.6 , yield 0.083 g (90%); m.p. 163-165°C; NMR (Figure 4.3) (DMS0d 6 ) 

6 8.90 (1H, d, J=6Hz), 7.82 (1H, d, J=6Hz), 7.25 (5H, s), 7.05 (2H, b), 

4.65 (2H, 2), 3.80 (3H, s); IR (KBr) cm' 1 3420, 3350, 1670, 1650, 1610, 

1560, 1220, 1010, 700. Anal. Calc. for C 17 H 14 3 N 2 -l/3 H 2 0: C, 68.0; 

H, 4.87; N, 9.33. Found: C, 68.02; H, 4.82; N, 9.30. 

N-[2-(2,3-Dimethoxy-5-benzy!oxypheny1 )-2-methoxyethyl]-2'nitropheny1 - 
acetamide 4.7 

A stirred suspension of 2.4 (1.3 g) and triethylamine (1.5 g, 3 
eq) in dichloromethane (50 mL) was treated dropwise with o-nitrophenyl- 
acetyl chloride until TLC indicated the absence of the phenyl ethyl amine. 
Water was added and the reaction mixture stirred for 30 min to decom- 
pose the excess acid chloride. The mixture was diluted with ethyl ether 
and the ether layer washed twice with aqueous acid and aqueous sodium 
bicarbonate, dried and concentrated. The product, a beige crystalline 
solid was recrystallized from ethyl ether/1 i groin, 4.7 , yield 1.4 g 
(93%) m.p. 95-97°C; NMR (CDC1 3 ) 6 8.05 (1H, m), 7.62- 7.26 (8H, m), 
6.89 (2H, s), 5.04 (2H, s), 4.69 (0.5H, d, J=4.5Hz), 4.59 (0.5H, d, 
J=4.5Hz), 3.3 (1H, d, J=4.5Hz), 3.23 (3H, s); IR (KBr) cm" 1 3380, 2940, 
1680, 1590, 1520, 1490, 1350, 1140. Anal. Calc. for C 26 H 28 7 N 2 : C, 
65.0; H, 5.8; N, 5.8. Found: C, 64.71; H, 5.92; N, 5.77. 



l-(2'Nii 
l-(2'nii 



i trobenzyl ) -5 ,6-dimethoxy-8-benzy1 oxyi soqui no! i ne 4.8 and 

Ftrobenzyl ) -4 , 5 ,6-trimethoxy-8-benzyl oxy-3 ,4-di hydroi sogui no! i ne 
4.9 



To a solution of 4.7 (1.5 g) in benzene (50 mL) was added phos- 
phorus oxychloride and the reaction mixture stirred under reflux 
overnight. TLC indicated the absence of starting material and the 
presence of two basic compounds. The cooled solution was poured over 
ice to decompose the excess POCI3 and stirred until all the ice had 






89 



ru 




n 



IS 



KO 



C 
:: 
O 

I 



E 

ZJ 

4-> 
U 

ex 

01 



01 



"01 



90 

melted. The acidic aqueous mixture was washed twice with ethyl ether, 
made basic with cold 20% sodium hydroxide and extracted with chloro- 
form. The chloroform layer was dried and concentrated to give 1.3 g of 
a mixture of 4.8 and 4.9 , which were used without further purification 
in the next step. A portion of this mixture was chromatographed over 
alumina in benzene to provide analytical samples of 4.8 and 4.9 ; 4.8 
eluted with benzene, a greenish-yellow crystalline solid (ethyl ether); 
m.p. 143-145°C; NMR (CDC1 3 ) 6 8.36 (1H, d, J=6Hz), 7.96 (1H, m), 7.79 
(1H, d, J=6Hz), 7.46-6.89 (8H, m) , 6.71 (1H, s), 5.16 (2H, s), 5.10 
(2H, s), 3.92 (6H, s); IR (KBr) cm" 1 3070, 2960, 1605, 1555, 1515, 
1350, 1230, 1120, 1060, 1000, 750, 720. 

4. 9 . --Eluted with 2% acetone in benzene, converted to the chlor- 
ide salt and crystallized from acetone/ethyl ether; m.p. 175-185°C 
(loss of HC1); NMR (DMS0d6) (poorly resolved); 6 8.09 (1H, m), 7.66 
(2H, m), 7.26 (6H, m), 6.98 (1H, s), 5.23 (2H, bs), 4.92 (3H, b), 4.36 
(1H, s), 4.18 (1H, s), 3.99 (3H, s), 3.81 (3H, s), 3.33 (3H, s); IR 
(KBr) cm" 1 2950, 2840, 1630, 1590, 1520, 1340, 1280, 1250, 1240, 
1200, 1045, 1000. 
l-( 2 ' Ni trobenzyl ) -5 ,6-dimethoxy-8-hydroxyi soqui no! i ne 4.10 

Conversion of the mixture of 4.8 and 4^9 (1.7 g) to 4.10 was car- 
ried out as under 4.3 . The product, a tan crystalline solid was crys- 
tallized from ethyl ether/1 igroin, yield 0.9 g (72%); m.p. 123-125°C; 
NMR (DMS0d 6 ) 6 8.14 (1H, m), 6.89 (1H, s), 5.11 (2H, s), 3.92 (3H, 
s), 3.80 (3H, s); IR (KBr) cm" 1 3100, 1610, 1560, 1515, 1350, 1260, 
1200, 1150, 820. Anal. Calc. for C 18 H 16 5 N 2 'H 2 0: C, 60.33; H, 5.02; 
N, 7.82. Found: C, 60.59; H, 5.03; N, 7.82. 



91 

l-(2'Nitrobenzy1 )-6-methoxy-7-bromoisoquino1ine-5 > 8-dione 4.11 

The procedure used was the same as that described under 4.4 ; 
from 0.377 g, yield 0.408 g (91%) as a dark yellow-brown crystalline 
solid from ethyl ether; m.p. 153-155°C; NMR (CDCI3) 6 8.73(1H, d, 
J=6Hz), 8.26 (1H, dd, J=7.5 + 1.5Hz), 7.79 (1H, d, J=6Hz), 7.51 (2H, 
td, J=7.5 + 1.5Hz), 7.31 (1H, dd, J=7.5 + 1.5Hz), 5.12 (2H, s), 4.30 
(3H, s); IR (KBr) cm" 1 3095, 1670, 1650, 1570, 1510, 1335, 1230, 
1040, 780, 740, 725, 715. Anal. Calc. for C 17 H n 5 N 2 Br: C, 50.7; 
H, 2.7; N, 6.96. Found: C, 50.71; H, 2.75; N, 6.91. 
l-(2'Nitrobenzyl)-6-methoxy-7-azidoisoquinoline-5,8-dione 4.12 

A procedure was used similar to that described under £^5 from 
0.155 g of 4.11 and 0.027 g (1.1 eq) of sodium azide, yield 0.12 g 
(85%) as a red-brown crystalline solid from ethyl ether; m.p. 115°C 
dec; NMR (CDCI3) 6 8.76 (1H, d, J=6Hz), 8.11 (1H, dd, J=7.5 + 
1.5Hz), 7.83 (1H, d, J=6Hz), 7.56 (2H, td, J=7.5 + 1.5Hz), 7.36 (1H, 
dd, J=7.5 + 1.5Hz), 5.12 (2H, s), 4.73 (3H, s); IR (KBr) cm" 1 
3095, 2120, 1660, 1590, 1520, 1340, 1300, 1240, 1080, 860, 840, 730. 
l-(2'Nitrobenzyl)-6-methoxy-7-aminoisoguinoline-5,8-dione 4.13 

Conversion of 4.12 (0.1 g) to 4.13 was carried out as described 
under 4.6 . The product was chromatographed over alumina in chloro- 
form and the major band eluted with chloroform, to give from ethyl 
ether a shiny dark red crystalline solid, yield 0.083 g (90%); m.p. 
237-240°C; NMR (Figure 4.4) (DMS0d 6 ) 6 7.61 (2H, td, J=7.5 + 
1.5Hz), 7.37 (1H, dd, J=7.5 + 1.5Hz), 7.06 (2H, b), 4.97 (2H, s), 
3.83 (3H, s); IR (KBr) cm" 1 3480, 3380, 1670, 1640, 1600, 1550, 
1520, 1350, 1235. Anal. Calc. for C 17 H 13 5 N 3 -l/3 H 2 0: C, 59.13; 
H, 3.94; N, 12.17. Found: C, 59.08; H, 3.75; N, 12.13. 



92 



f\i 




m 



CO 



■o 

c 

o 

Q. 

o 
o 



IS 



i. 
+J 
u 

O) 
Q. 



00 



0) 

CI 



-or 



93 

l-(2' Ami nobenzyl ) -6-methoxy-7-ami noi soqui nol i ne-5 ,8-di one 4.14 

To a solution of 4.12 (0.2 g) in THF (15 mL) water (15 ml_), 
methanol (2 mL) was added sodium dithionite (2 g) and the reaction 
mixture was stirred under reflux for 30 min. When TLC indicated the 
absence of starting material, the reaction mixture was made acidic 
with 50% sulfuric acid (3 mL) and reflux continued for an additional 
30 min. The cooled reaction mixture was filtered to remove precipi- 
tated sulfur, diluted with water and washed once with ethyl ether. 
The acidic aqueous layer was made slightly basic with cold 20% sodium 
hydroxide and extracted twice with ethyl acetate which was washed 
twice with water, dried and concentrated. The residue was placed on 
an alumina column in chloroform and the major band eluted with the 
same solvent to give from li groin, 4.14 as a dark red crystalline 
solid, yield 0.091 g (54%); m.p. sublime 135°, melt 198°C; NMR 
(Figure 4.5) 5 8.83 (1H, d, J=6Hz), 7.77 (1H, d, J=6Hz), 7.50 (4H, 
bm), 7.17 (2H, b), 4.03 (2H, s), 3.81 (3H, s); IR (KBr) cm" 1 
3405, 3320, 1620, 1590, 1520, 1430, 1240, 1040. Anal. Calc. for 
C 17 H 15 3 N 3 : C, 66.01; H, 4.85; N, 13.59. Found: C, 66.11; H, 4.34; 
N, 13.60. 



94 



-7 



L^ 




f\j 



m 



IB 



T3 

o 

Q. 

j= 
O 
O 

o 

s. 

o 
<v 

on 
Qi 



0) 
CD 



"W 



CHAPTER V 
METHYLAMINO AND DIMETHYLAMINO QUINONES 

Introduction 

As stated previously, the amino p-quinone of the A ring of 
streptonigrin was found to be essential for activity (43). Few 
changes could be made without loss of activity. A possibility here 
would be to replace the amino group with an alkylamino or diakylamino 
group. This would increase the lipid solubility and stabilize the 
amine to hydrolysis. The dialkylaminoquinone should be less potent 
than the alkyl ami noqui none due to a loss of hydrogen bonding abil- 
ity. 

In the 9-anilinoacridine antitumor compounds (62), which are 
quinonoid in structure, the alkylamino derivative was more potent 
than the amino derivative. The dialkylamino derivative was less 
potent than either of the two former compounds against L1210 leukemia 
but it was superior in solid tumor lines, such as Lewis lung carcin- 
oma. This was explained in terms of an increase in lipid solubility 
and higher uptake of the compound. It was decided to synthesize the 
methylamino and dimethylamino derivatives of some of the isoquinoline 
quinones previously presented. 

In this process, it was found that reactions of methyl amine 
with the 1-substituted 6-methoxy-7-bromoisoquinoline-5,8-dione did 
not give the expected 6-methoxy-7-methylaminoisoquinoline-5,8-dione 
but the 6-methylamino-7-bromoisoquinoline-5,8-dione. 

The following evidence suggested that an unexpected substitu- 
tion reaction had taken place: the absence of the methoxy signal 

95 



96 



in the NMR spectrum usually found at 6 4.0 - 3.8 ppm and the presence 
of an N-methyl signal, instead (Figure 5.1). The mass spectra of the 
compounds giving molecular ions were also consistent with the pro- 
posed structure. Although this is a common reaction in the mitosene, 
mitosane (63, 64) and quinoline (38, 39) quinones, it has not been 
reported in isoquinoline quinones. 

Similar results were also obtained with dimethylamine and pip- 
eridine. The reaction of piperidine with l-(4-nitrophenyl )-6-meth- 
oxy-7-bromoisoquinoline-5,8-dione gave a product; the NMR spectrum of 
this product (Figure 5.2) gave more unequivocal evidence in that it 
showed no methoxyl region but only the two multiplets at 6 1.78 ppm 
and 6 3.6 ppm which intergrated for 6 and 4 protons respectively. 

The behavior seemed unusual because reaction of the 1-substi- 
tuted 6-methoxy-7-bromoisoquinoline-5,8-dione with sodium azide gave 
the expected 6-methoxy-7-azidoisoquinoline-5,8-dione. This indicated 
possible involvement of a cyclic intermediate as seen in the quino- 
line quinone system (38, 39). It was therefore decided to explore 
this reaction more fully to find its limitations (Figure 5.3). Addi- 
tion of sodium azide to a solution of l-(4-nitrophenyl )-6-methyl- 
amino-7-bromoisoquinoline-5,8-dione ( 5.5 ) in DMF gave no reaction, as 
did treatment of l-(4-nitrophenyl )-6-methoxy-7-aminoisoquinoline-5,8- 
dione with methylamine. Treatment of l-(4-nitrophenyl )-6-methoxy-7- 
azidoisoquinoline-5,8-dione with methylamine produced the 6-methyl- 
amino-7-azidoquinone as evidenced by IR (Figure 5.4). This seemed to 
indicate that the group present in one position influences the 
reactivity of the other position. Those compounds with amino or 






97 



J 



— **w4UfL~-^v 



~\ m *i ■ l r -i t * in i ' 



I 1 



I 1 



— hMIXj ....oJ^ 



i — i- 






M>^|k 



t*l*nH Htmmmm*i0m 



HUP* mm ■Mn 



L i . J 



Figure 5.1. Comparison of 6-methoxy-7-aminoquinone with 
6-methyl ami no-1-bromoqui none. 



98 



in 



LT> 



-V 



-s 



1 





c 


< 

1 


o 

Q. 

5 
tj 


1- 


- vo o 


{ 


3 




o 

Q. 


!'- 


- r* §= 


£ 


■21 

• 


_n 


LO 

to 


^ 




j 





1 



99 




CH 



+ CH NH ^ > 



NO. 





NQN3 X > 



CH-N 




+ CH 3 NH 2 X > 



CH3 




+ CH 3 NH 2 



-> 




Figure 5.3. Unusual reactions. 






100 



poo 




5 

Q. 

o 
u 



3 

+-> 
o 



CD 



MOsawsNvit ifsrnj 



101 



methyl ami no groups do not readily undergo addition reaction; whereas 
those compounds with a methoxy at position 6 and a bromine or azide 
group at 7, lend themselves to addition reactions. Joseph and Joullie 
(52) found isoquinoline-5,8-dione underwent mono-addition with hydrogen 
chloride or aziridine. They postulated that addition took place at 
position 6 although they could not confirm this, but based it on analo- 
gy with a quinoline-5,8-dione system, there the 6-chloroisomer was pro- 
duced exclusively on treatment with hydrogen chloride followed by oxi- 
dation. 

In a recent report, Shaikh et al . (45) claimed that treatment of 
6,7-dichloroisoquinoline-5,8-dione with aqueous sodium hydroxide gave 
6-chloro-7-hydroxyisoquinoline-5,8-dione. This assignment was based 
mainly on a resonance structure which placed positive charge at the 7 
position. If this resonance structure with positive charge at 7 does 
have a large contribution, one would expect the reaction of methylamine 
with a 1-substitute 6-methoxy-7-bromoisoquinoline-5,8-dione to yield 
6-methoxy-7-methyl ami noqui none and not the 6-methyl ami no-7-bromoqui none 
as was seen in the present case. 

The structures of the compounds synthesized are presented in 
Figure 5.5. 

Experimental 

Nuclear magnetic resonance (NMR) spectra were recorded on a Vari- 
an EM 390 and chemical shifts (6) were reported in parts per million 
(ppm) relative to internal tetramethylsilane (0.0 ppm). Infrared (IR) 
spectra were recorded on a Beckman Acculab 3 as KBr pellets and report- 
ed as cm" 1 . Melting points were determined on a Fisher-Johns hot 
stage melting point apparatus and are uncorrected. Elemental analysis 
was carried out by Atlantic Microlabs, Atlanta, GA. Alumina (Woelm) 
was used at activity grade III. 



102 




CH- 






NO, 




CH- 




ar^ 





p> 



CH, 




3r- 



^ ,v 



n 



Figure 5.5. Structure of compounds synthesized. 



103 

l-(4 ' Ni trophenyl ) -6-( 1-pi peri di nyl ) -7-bromoi soqui nol i ne-5 ,8-di one 5.1 

To a stirred solution of 3A_ (0.1 g) in DMF (10 mL) was added 
excess pi peri dine (0.1 mL). The solution immediately turned dark 
purple and, after 15 min, TLC indicated the absence of starting 
material. The reaction mixture was diluted with water and extracted 
with ethyl acetate. The ethyl acetate layer was washed with dilute 
aqueous acid and water, dried and concentrated. The product was crys- 
tallized from ethyl ether/ligroin as a violet crystalline solid, 5.1 , 
yield 0.095 g (83%); m.p. 168-170°C; NMR (CDC1 3 ) 6 9.06 (1H, d, 
J=6Hz), 8.35 (2H, d, J=9Hz), 8.0 (1H, d, J=6Hz), 7.63 (2H, d, J=9Hz), 
3.6 (4H, m), 1.78 (6H, m); IR (KBr) cm" 1 2950, 2860, 1670, 1640, 
1570, 1540, 1510, 1340. Anal. Calcd. for C 20 H 16 4 N 3 Br: C, 54.31; H, 
3.65; N, 9.50. Found: C, 54.25; H, 3.69; N, 9.45. 
l-( 4 'Ni trophenyl ) -6-methyl ami no-7-azidoi soqui noli ne-5, 8-di one 5.2 

To a solution of 3.5 (0.05 g) in DMF (10 mL) was added excess 
40% aqueous methylamine (0.2 mL) and the reaction mixture was 
stirred. On the addition of the methylamine the reaction mixture 
turned dark blue and TLC indicated the absence of starting materials. 
The reaction mixture was diluted with water and the solid filtered. 
The solid was crystallized from THF as a dark blue solid, 5.2 , yield 
0.025 g (50%); m.p. 200°C dec; IR (KBr) cm" 1 3360, 2120, 1665, 
1600, 1510, 1340. 

General procedure for the synthesis of the methyl amino com - 
pounds . --To a solution of the 6-methoxy-7-bromoi soqui noli ne-5, 8-d i one 

in DMF was added excess 40% aqueous methylamine and the reaction 

mixture was stirred until TLC indicated the absence of starting 



104 



material. The reaction mixture was diluted with water and extracted 

with ethyl acetate. The ethyl acetate layer was washed with dilute 

aqueous acid and with water, dried and concentrated. The residue was 

placed on an alumina column in chloroform and eluted with the same 

solvent. 

1-Benzyl -6-methyl ami no-7-bromoi soqui nol i ne-5 ,8-di one 5.3 

The product, 5^3, yield 0.085 g (85%) from 0.1 g of 4^4, was 
crystallized from ethyl ether/ligroin as a bright red solid; m.p. 
95-98°C; NMR (Figure 5.6) (DMS0d 6 ) 6 8.87 (1H, d, J=6Hz) , 7.75 (1H, 
d, J=6Hz), 7.23 (5H, s), 4.75 (2H, s), 3.23 (3H, s); IR (KBr) cm -1 
3360, 1675, 1640, 1600, 1520, 1300, 1110, 700, 690. Anal. Calc. for 
C 17 H 13 2 N 2 Br: C, 57.30; H, 3.65; N, 7.86. Found: C, 57.24; H, 
3.65; N, 7.82. 
l-( 2 ' Ni trophenyl ) -6-methyl ami no-7-bromoi soqui nol i ne-5 ,8-di one 5.4 

The product, 5.4 , yield 0.075 g (75%) from 0.1 g of 3.22 was 
crystallized from ethyl ether as a dark red solid; m.p. 195°C, dec; 
NMR (Figure 5.7) (DMS0d 6 )6 8.99 (1H, d, J=6Hz)., 8.23 (1H, dd, 
J=7.5 + 1.5Hz), 7.97 (1H, d, J=6Hz), 7.83 (2H, td, J=7.5 + 1.5Hz), 
7.4 (1H, dd, J=7.5 + 1.5Hz), 3.3 (3H, s); IR (KBr) cnr* 3360, 
1670, 1590, 1510, 1340, 740. Anal.: Unstable does not give good 
analysis. 
l-( 4 'Ni trophenyl ) -6-methyl ami no-7-bromoi soqui noli ne-5, 8-di one 5.5 

The product, 5.5 , yield, 0.08 g (80%) from 0.1 g of 3A was 
crystallized from ethyl ether/ligroin as a dull red solid; m.p. sub- 
lime at 245°C (melts 255°C) ; NMR (Figure 5.8) (DMS0d 6 ) 5 9.0 (1H, 
m), 8.3 (2H, d, J=9Hz), 7.97 (1H, m), 7.67 (2H, d, J=9Hz) , 3.3 (3H, 
s); IR (KBr) cm" 1 3360, 1670, 1640, 1600, 1500, 1340, 740. Anal. 
Calc. for C 16 H 10 4 N 3 Br: C, 49.5; H, 2.58; N, 10.8. Found: C, 
49.57; H, 2.65; N, 10.77. 



105 




t\j 



m 



en 

-a 
c 

o 
a. 

o 

u 



vS 




u 

CD 

a. 
v> 



'JD 



s_ 

ZJ 



"CI 



106 




1 



•- o 



in 



C 
S 

o 
a, 

5 
u 



e 
i. 

u 
a. 



s_ 

3 

en 



a\ 



107 




n 



in 



\D 



CO 



in 
c 

3 
O 
2. 

e= 

o 



E 

3 
S. 
4-> 

u 

tu 
a. 

IS) 



ai 
en 






108 



l-(3' Nitrophenyl )-6-nethylamino-7-bromoisoguinol ine-5,8- 
dione 5.6 

The product, 5.6 , yield, 0.09 g (90%) from 0.1 g 3.13 was 

crystallized from ethyl ether as a red solid; m.p. 250°C dec; NMR 

(Figure 5.9) (DMS0d 6 ) 6 9. 03 (1H, d, J=6Hz), 8.3 (2H, m), 7.97 (1H, 

d, J=6Hz), 7.76 (2H, m), 3.26 (3H, s); IR (KBr) cm" 1 3360, 1670, 

1600, 1520, 1340, 740, 670. Anal. Calc. for C 16 H 10 4 N 3 Br: C, 49.6; 

H, 2.58; N, 10.8. Found: C, 49.58; H, 2.63; N, 10.81. 

l-(2' Nitrophenyl )-6-dimethylamino-7-bromoisoquino1 ine-5,8- 
dione 5.7 

To a solution of 3.22 (0.1 g) in DMF (10 mL) was added 25% 

aqueous dimethyl amine (0.5 mL) and the reaction mixture stirred until 

TLC indicated the absence of starting material. The reaction mixture 

was diluted with water and extracted with ethyl acetate. The ethyl 

acetate layer was washed with dilute aqueous acid, water, dried and 

concentrated. The residue was chromatographed over alumina in 

chloroform and eluted with the same solvent. The product, 5.7 , was 

crystallized from ligroin/dichloromethane to give a dark purple 

solid, yield, 0.08 g (77%); m.p. 127°C, dec; NMR (Figure 5.10) 

(CDCI3) 6 8.97 (1H, d, J=6Hz), 8.26 (1H, dd, J=7.5 + 1.5Hz), 7.93 

(1H, d, J=6Hz), 7.67 (2H, td, J=7.5 + 1.5Hz), 7.33 (1H, dd, J=7.5 + 

1.5Hz), 3.23 (6H, s); IR (KBr) cm" 1 1670, 1630, 1570, 1510, 1340, 

1210, 740. Anal. Calc. for C 17 H 12 4 N 3 Br: C, 50.87; H, 3.0; N, 

10.47. Found: C, 50.83; H, 3.06; N, 10.45. 






109 



<M 



i. 



n 




\Ti 



■' — 



vO 




ir> 

c 

Z3 
O 
Q. 

E 
O 



+-• 

O) 



in 



no 



-4 




CHAPTER VI 
RESULTS AND DISCUSSION 

Introduction 

The biological activity of the isoquinoline derivatives was 
evaluated using two in vitro methods, the disk-plate antibiotic assay 
against _B. subtil is and the root growth inhibition assay using cress 
seedlings ( Lepidium sativum ) (65). Comparison of the isoquinoline 
derivatives with the quinoline derivatimes in the same test systems 
shows the relative potency of these compounds. These assays also 
show the effect of changes in the C ring within the isoquinoline 
series. 

A plant seedling assay has many advantages over other test sys- 
tems, such as mammalian cell cultures, in that it is inexpensive; 
takes little time to set up or prepare and requires no special equip- 
ment other than possibly an incubator. The rationale for the assay 
as a screen for potential antitumor agents lies in the fact that the 
apical meristem of the root is composed of actively dividing, undif- 
ferentiated cells. This description is similar to that of cancer 
cells. This is not a general assay; compounds that affect animal 
enzyme systems may not affect the plant's growth and conversely, com- 
pounds that affect plant systems may not affect animal cells. For 
compounds to show activity in this assay the test compound must be 
absorbed by the plant via the root system. This requires that the 
compound have some degree of water solubility. Compounds with lim- 
ited water solubility may show results which are inconsistant with 

111 



112 



those from other compounds in the same series. This problem arises 
in the case of some of the nitro-substituted compounds which will be 
discussed later. 

The structures of the compounds tested are presented in Figure 

6.1. 

Material and Methods 
Root Growth Inhibition Assay 

Cress seeds ( Lepidium sativum ), soaked in water for 15-20 min, 
were placed (2 seed/square) on 0.6% agar in phage typing petri dishes 
(Falcon) and were allowed to germinate in the dark at 27°C. After 24 
h, the radicals were approximately 1 cm long. 

The compounds to be tested were dissolved in DMSO and dilutions 
were made so that 0.2 mL of the test solution when diluted with 20 mL 
of hot 0.6% agar, would give concentration of 20, 10, 4, 2, 0.8 and 
0.4 yg/mL. The hot agar solutions were poured into the square phage 
typing dishes and were allowed to cool. 

Those seedlings with 1 cm radicals were placed on the test agar 
plates so that the root tips were at the intersection of two grid 
lines (8 seedlings/ concentration). Controls were 0.2 mL of DMSO/20 
mL agar. The plates were placed flat in an incubator at 27°C, after 
24 h the root lengths were measured and averaged to give an average 
root length for test concentrations (T) and control .(C). Each com- 
pound was run in duplicate and the data for the two runs were com- 
bined. Percent inhibition (%I) was calculated using the formula: 
1 - (T/C) x 100 = %I. 

The data were analyzed using probit analysis (66) in which per- 
cent value is converted to probability using a table of probit. The 
data were subjected to linear regression using a calculator program. 



113 




CH 




h-cx^An 




nr 



K^ 



4.6 




CH 




CH 3 



H 2 N 






'K*fl 



3.6 





CH 




Figure 6.1. Structures of compounds tested in the root 
growth inhibit assay and antibiotic assay. 



114 



H^N' 




CH 






>COOH CH 3 



dSN 



h 2 n^^^S:h3 




isoPyQ 



^^ 



CH 







SN 




)H 



CH- 



CH- 



Figure 6.1. Continued. 



115 



The concentration at which 50% inhibition of growth occurs 
(I50) was determined using the same program. Standard errors and 
95% confidence limits were calculated as per Finney (66). 
Antibiotic Activity Test 

Compounds to be tested were dissolved in DMSO at concentrations 
such that 50 yl contained 100, 50, 20, 10, 4 or 2 ug. To antibiotic 
filter paper disks (Schleicher and Schuell 12.5 mm) was added 50 yl 
of the test solution. The disks were placed equally spaced on 
antibiotic test agar plates containing j3. subtil is spores and the 
plates were placed in an incubator at 37°C. After 24 h, the zones of 
bacterial inhibition were measured and the results from 4 replicates 
were combined. The data were subjected to linear regression using a 
calculator program. A plot of the diameter of zone of inhibition 
versus logig of the concentration is a linear relationship. The 
concentration at a zone diameter of 13 mm was taken as the minimum 
inhibitory concentration (mic). Standard errors for the slope were 
calculated as per Strike (67). 

Results 
The Root Growth Inhibition Assay (Figures 6.2 - 6.6) 

Comparison of the I50 values (Table I) in the root growth 
inhibition assay for the l-(2-pyridyl )isoquinoline, 2.17 ; phenyl iso- 
quinoline, 2.11 ; benzyl isoquinoline, 4.6 ; 6-amino-7-methoxy-2(2- 
pyridyl )quinoline (isoPyQ) and streptonigrin (SN) indicate only 
slight differences in their potencies (isoPyQ > 2.11 > SN > 2.17 > 
4.6 ). This is not a very surprising result since data generated by 
Boger et al . (44) indicate that streptonigrin and the 6-methoxy-7- 
amino-2-(2-pyridyl )quinoline derivative are of approximately equal 



116 



". ". z 

CN CN VI 



id 



m 

x 



o 
tr 

CJ3 

I— 
O 

o 

Of 




in 

CN 

cn 



CN 
0> 



O 
co «- 



21 

- 3 

- 2 



• UJ 

- o 
z 

- o 
o 



.. a> 



.. a 



.. p» 



.. « 



.. i*> 



oo 



S- 
Cl 

o 
+-> 

^- 

OJ 

s_ 

00 

c 



CM 



CM 

00 

O 
Ql 

E 
o 
u 



o 

00 



Q. 



O 



(XI 



S- 
=3 
CD 



PROBIT 



117 






UJ 

o 

UJ 



CN 



z 
1/1 



I 



z 
g 

m 

x 



o 

K 

I— 
O 

o 




rj 

CN 



CN 



CN 
CN 

at 



2s 

: ? 

n Z 
- 2 

si 



- o 
z 

- o 
u 



.. at 



.. o 



.. r>» 



.. in 



.. r> 



.. CN 



U~) 



en 
■t — 

c 

o 
+-J 

Q. 

O) 

s_ 
■g 

CM 

• 

CO 

I— t 
CO 

r-«| 

• l 

CO I 

</) 

o 

Q. 

O 

o 



o 

•r- 

s- 

A3 

Q. 

S 



CO 

C71 



<n 



PROBIT 



118 



a 
z 

UJ 

O 






i/i 



m 



o 
a: 

t- 
o 
o 

(S 










■ • c« 

a 

oq 



O 



" ~ g 
■■ ^ a: 



z 

- o 

o 



" °! 
•• "J 



a 
o 



— H 1 1- 

10 <r n 

PROBIT 



CO 



cr, 

•r— 

o 
+-> 
c 
ai 
s- 
■t-> 
i/> 

"O 
CM 



in 

i— i 

• 

ro 

CO 

■a 

o 

Q. 

E 

o 
o 



c 
o 

z. 

Q. 
E 
O 

o 



"3- 



ai 



o> 



119 




S- 

C 
O 

+-> 

a. 
01 

s_ 
+-> 

"O 

c 
"3- 



CNJ 

l/l 

TJ 

C 
3 
O 
CL 
C 

o 



c 
o 

CO 
•r- 

S- 

ra 

CL 
E 

o 



10 

3 
C7> 



120 







o 
a. 

CD 

u 
+J 
to 



o 

JE 
Q. 

X 

o 



1/1 

01 

-o 



cn 

c 
o 

Q. 



r- >> 

.—I Q. 

. O 

CM (/5 

T3^— 

C 

3 cu 
o c 
a. o 

E = 
o->- 

O 3 

CT 
<4- O 
O C 

c'e 

O fO 

</1 

•i- 0) 

s- c 

(O -i- 
Q.I— 
E O 
O C 

3 
CT 

<£> >-> 
.-o 

UD •■- 

{_ 

CU >5 

S- Q- 

3 

a>-o 

•f- c 

Li_ ro 



T3 



121 







ROOT 


TABLE I 
INHIBITION 


ASSAY 








150 




95% Confidence 
Units in nM/mL 


Relative 
Potency to 
Streptonigrin 




Compound 


ug/mL 


nM/mL 


r 


2.11 


3.15 


11.25 


7.46 - 


17.09 


1.12 


.99 


2.17 


3.57 


12.7 


8.15 - 


19.5 


0.99 


.97 


4.6 


4.05 


13.7 


9.15 - 


20.97 


0.92 


.97 


3.7 


4.57 


15.49 


10.05 - 


24.0 


0.82 


.95 


3.16 


6.6 


22.37 


15.85 - 


31.6 


0.57 


.99 


3.25 


19.26 


65.0 


32.37 - 


125.94 


0.19 


.98 


3.6 


3.06 


9.41 


6.43 - 


14.06 


1.34 


.97 


3.15 


2.09 


6.43 


4.77 - 


8.68 


1.96 


.98 


3.24 


1.96 


6.03 


4.55 - 


7.91 


2.09 


.99 


Streptonigrin 


6.4 


12.65 


8.23 - 


19.76 


1.0 


.98 


dSN a 


1.95 


5.5 


3.57 - 


8.53 


2.3 


.99 


i soPyQ b 


2.69 


9.18 


6.11 - 


11.22 


1.37 


.94 


Eu3 


10.7 


30.05 


24.26 - 


36.72 


0.42 


.96 


5^4 


3.86 


10.0 


7.76 - 


12.30 


1.3 


.94 


3.8 




>65 


- 




- 




3.17 




>65 


- 




- 





a dSN = destrioxyphenyl streptonigrin. 

b i soPyQ = 6-ami no-7-methoxy-2- ( 2-pyri dyl )qui no! i ne-5 ,8-di one. 



122 

potency in L1210 cell culture (I 50 SN, 1.2 nM; PyQ, 1.06 nM; potency 
ratio 1.12). This comparison does not hold true for all cell cul- 
tures tested in this study. In some cell lines, streptonigrin was 
substantially more active than the pyridylquinoline whereas in 
others, the pyridylquinoline derivative was more potent. The 
differences in potencies are possibly attributable to differences in 
cellular uptake of the compounds. The sensitivity of the cell lines 
could also be a factor. 

Destrioxyphenyl streptonigrin (dSN) was found to be more than 
twice as potent as streptonigrin in the root growth inhibition assay. 
The increase in potency of this compound could be attributable to the 
lack of the dimethoxyphenol substituent, which would impart more lip- 
id character to streptonigrin and decrease its water solubility and 
thus uptake by the plant. 

The aminophenylisoquinoline qui none derivatives were, as a 
whole, less potent than the phenyl isoquinoline quinone. Within the 
group, the p-aminophenyl derivative with an I50 of 15 nM/mL was 
the most potent, followed by the m-aminophenyl with an I50 of 22 
nM/mL. The o-aminophenyl derivative was the least potent of the 
group with an I5Q of 65 nM/mL. The acetamides of the para and 
meta aminophenyl derivatives were much weaker with I50 values 
greater than 65 nM/mL. 

The nitrophenyl isoquinoline qui nones were found to be more ac- 
tive than the phenyl isoquinoline quinone; the o-nitro and m-nitro- 
phenyl derivatives with I50 values of 6 nM/mL and 6.5 nM/mL 



123 



respectively, were equi potent. The decrease in potency observed for 
the p-nitro compound (I50, 9 nM/mL) was due to problems with its 
insolubility in water. 

The methylamino derivatives showed only weak activity in the 
plant assay due to their low water solubility. Only the methylamino 
o-nitrophenylisoquinoline qui none 5.4 with an I50 of 10 nM/mL and 
the methylamino benzyl isoquinoline quinone 5.3 with an I50 of 30 
nM/mL showed any activity. The dimethyl ami no derivatives were not 
active at the levels tested (>20 ug/mL). 
Antibiotic Assay (Figures 6.7 - 6.11) 

The isoquinoline derivatives were all less potent than strepto- 
nigrin in this assay (Table II). A comparison of the pyridyl isoquin- 
oline , phenyl isoquinoline and benzyl isoquinoline qui nones showed that 
the benzyl isoquinolines (mic 16 nM) to be the most active of the 
group, followed by the phenyl isoquinoline (mic 27 nM) and the pyrid- 
ylisoquinoline which was the least active (mic 39 nM). 

In contrast to the activity seen in the root inhibition, the 
o-aminophenyl derivative was 1.5 times as active as the unsubstituted 
phenyl isoquinoline quinone whereas the p-aminophenyl (0.70 times) and 
the m-aminophenyl (0.80 times) derivatives were less active. In the 
nitrophenyl series the o-nitro and m-nitrophenyl derivatives were 3 
and 5 times respectively, as active as the unsubstituted phenyl 
derivative. The p-nitrophenyl compound was too insoluble in water to 
diffuse. 

The methylamino and dimethylamino derivatives were, in general, 
too insoluble in water to diffuse into the agar; only two, the meth- 
ylamino benzyl isoquinoline, 5.3 , and the methylamino o-nitrophenyl- 
isoquinoline, 5.4 , derivatives gave measurable zones. The methyl- 
amino benzyl isoquinoline quinone was 4 times as potent as the 






124 



2 -. % 

01 CN Q. 

a ° 

i * 

Ui X Q 

o 



> 


jn 


H- 


■_^ 


a j; 


< 


3 




M 


_i 




< 


oa 


cr 




UJ 


h- 


h- 


on 


G 


7" 


< 




m 


<. 




<J 


h- 


< 




~ Z 
C 

- < 



CJ 
o 2 
- O 

CJ 



.. Oft 



.. a 



.. r*» 



.. <o 



.. « 



.. n 






a 



in 

CN 



a 

CN 



ZONE DIAMETER (mm) 






3 

>i 

-a 

'i — 

s- 

■a 

c 

03 



o> 

*r— 

o 

Q. 
0> 

s. 
+-> 

CO 



C 

O 

a. 

§ 



o 

CO 

•r— 

S_ 
<T3 
Q. 

E • 

O— • 

OCT 

>> 

a. 

. o 
r-~ co 

• 'r* 

io> — 

<u <u 
s_ c 
3 o 

Li_ 3 

a - 

o 
e 



125 



2 -. «. 
tfj W » 




- < 



- u 

o 2 

-• o 



ZONE DIAMETER (mm) 



c/o 





s_ 




o> 




•^ 


^ 


e 




o 




4-> 




a. 


•- ^™ , 


at 


o> 


S- 


n = 


to 


•" Z 


"O 


o 


e 


«. f= 


(a 



to 



to 

"O 

c 

3 
O 

a. 

s= 
o 
u 



c 
o 

tsi 

•r- 

s_ 

(O 

a. 

E 
C 

o 



cc 



126 




c 

•r— 



c 
o 
+-> 

Q. 
11 
[_ 

4-> 

"O 

C 
<T3 

Lf) 

CM 

• 



.—I 

CO 



ro| 
1/1 

o 

a. 

b 

o 

«t- 

o 

c 
o 

•r- 

s_ 

Q. 

E 
o 
o 



* 

s_ 
3 
en 



ZONE DIAMETER (mm) 



127 



z *"• £ 
c/> rj n 








S- 


CO 




o> 






•r— 


*" 




c 






o 


m 




4-> 






Q. 






s- 

-t-> 






(/) 


*— 


,»-*t 






o> 


T3 


n 


3 


c 


*■ 


z 




CN 


o 


"=t 


t— 


C\J 


•■ 


< 


• 




K 


CO 



o Z 

- o 



ZONE DIAMETER (mm) 



LO 



oo| 

C 
3 
O 
Q. 

O 
CJ 



c 
o 
co 

•r— 

S- 
(0 
Q. 

O 

c_> 



u 



128 



Z i»5 «0 

m iri * 



Ui X c 

a 

UJ 



> 


05 


1— 


™ 


u 


£ 


< 


3 




en 


_i 




< 


rn 


IX 




LU 


i— 


I— 


t/i 


u 


Z 


-J 




03 


<- 


>— 


< 






o 











o 



1(1 






Si 



o Z I 

- c ! 

o 
_j 

.. a 



.. m 



.. n 



ZONE DIAMETER (mm) 



l/l 



s_ 



o 
+-> 
a. 
<u 

S- 

+J 
t/> 

-o 

c 

<T3 
CO 

* 

n 

• 

to 

-a 

c 

ZJ 
O 
Q. 
£= 
O 

o 

4- 
O 

C 

o 

<T5 
Ol 

E 
o 
o 



CD 



129 







TABLE 
ANTIBIOTIC 


II 
ASSAY 














mic 




Slope : 


t SE 




Compound 


yg 




nM 




r 


2.11 


7.63 




27.25 




14.78 


+ 


0.801 


0.9999 


2.17 


11.03 




39.25 




12.37 


+ 


0.908 


0.973 


4.6 


4.72 




16.05 




11.31 


+ 


1.4 


0.932 


3.7 


10.54 




35.8 




12.64 


+ 


1.1 


0.99 


3.16 


9.85 




33.4 




11.7 


+ 


0.817 


0.984 


3.25 


5.07 




17.2 




9.20 


+ 


0.914 


0.97 


3g6 


No diffus 


ion 


No diffusion 










3.15 


1.69 




5.2 




4.93 


+ 


0.50 


0.996 


3.24 


2.93 




9.0 




10.3 


+ 


0.80 


0.976 


Streptonigrin 


0.693 




1.37 




5.51 


+ 


0.226 


0.971 


dSN a 


0.088 




0.25 




4.75 


+ 


0.26 


0.99 


i soPyQ b 


3.97 




14.13 




10.23 


+ 


0.91 


0.99 


5.J3 


4.72 




16.0 




11.31 


+ 


1.3 


0.932 


5.4 


2.4 




6.76 




5.21 


+ 


0.45 


0.977 


3.8 


No diffusion 


No diffusion 










3.17 


No diffus 


ion 


No diffusion 











a dSN = destrioxyphenyl streptonigrin. 

b isoPyQ = 6-amino-7-methoxy-2-(2-pyridyl )quinoline-5,8-dione. 



130 

phenyl isoqui no! ine qui none and the methyl ami no o-nitro compound was 3 
times as potent. Thus, the methyl ami no and the amino o-nitrophenyl- 
isoquinoline derivatives appear to be approximately equipotent. 

Conclusions 
The Assays 

Comparison between the root growth inhibition assay and the 
antibiotic assay is not quite meaningful because they may be indicat- 
ing different types of activity. For example, in the root growth 
assay, streptonigrin was approximately equipotent to the phenyl iso- 
quinoline, benzyl isoquinoline and pyridylisoquinoline quinones, 
whereas in the antibiotic assay these compounds were extremely weak 
in comparison to streptonigrin. 

In the amino substituted compounds, the order of activity for 
the plant assay was p-amino > m-amino > o-amino, but in the antibi- 
otic assay this order was reversed. In the nitro series, the same 
situation exists; in the root growth assay, the o-nitro and m-nitro 
derivatives were approximately equipotent but in the antibiotic assay 
the m-nitro derivative was 1.5 times as potent as the o-nitro. These 
differences are not readily explained in terms of electronic effects 
or favorable hydrogen bonding interactions (Chapter III). The best 
explanation for these differences is based on differences in absorp- 
tion, cellular uptake or penetration. 

The root growth inhibition assay seems to be a good first 
screen for the selection of compounds that may interact with DNA. 
This assay should be studied more extensively with antitumor com- 
pounds having various mechanisms of action to examine how they 
react. 



131 

Further Research 

The isoquinoline analogues must be tested in vivo to truly 
evaluate their potential as antitumor agents. They also must be 
evaluated against the retrovirus, HTLV III; based on other work 
conducted in this area (30, 31), these compounds have a good chance 
of being active. 

Further research in the isoquinoline series of streptonigrin 
analogues should focus on the synthesis of the analogue in which the 
C-ring carries the same substituents as that of streptonigrin. This 
compound must be made and evaluated in comparison to streptonigrin 
and destrioxyphenyl streptonigrin before the potency of the 
isoquinoline ring system can be judged. 

The next heterocyclic ring system in logical progression from 
isoquinoline would be quinoxaline. This ring system has the best 
potential to produce an active series of analogues. Quinoxaline 
analogues would retain the metal binding properties of the quinoline 
system but its reduction potential should be different. 



REFERENCES 

1. Rao, K. V.; Cullen, W.P. Antibiot. Ann. 1959 , 950 (1959). 

2. Davis, H.L.; Von Hoff, D.D.; Henney, J.T. ; Rozencweig, M. 
Cancer Chemother. Pharmacol. _1, 83 (1978). 

3. Oleson, J.J.; Calderella, L.A.; Mjos, K.J.; Reith, A.R.; Thie, 
R.S.; Toplin, T. Antibiot. Chemother. U, 159 (1961). 

4. Reilly, H.C.; Sugiura, K. Antibiot. Chemother. _11_, 174 
(1961). 

5. Chirigos, M.A.; Pearson, J.W. ; Papas, T.S.; Woods, W.A.; Wood, 
H.B.; Spahn, G. Cancer Chemother. Rep. 57, 305 (1973). 

6. Inouye, Y.; Take, Y.; Nakamura, S. J. Antibiotics. 40, 100 
(1987). 

7. Wilson, W.L.; Labra, C; Barrist, E. Antibiot. Chemother. U, 
147 (1961). 

8. Hackethal , C.A.; Golbey, R.B.; Tan, C.T.C.; Karnofsky, D.A. ; 
Burchenal , J.H. Antibiot. Chemother. 11, 179 (1961). 

9. Pratt, W.B. and Ruddon, R.W. , The Anticancer Drugs . Oxford 
University Press, New York, p. 155, 165, 1979. 

10. Kaung, D.T.; Whittington, R.M.; Spencer, H.H.; Patno, M.E. 
Cancer. 23, 597 (1969). 

11. Kaung, D.T.; Whittington, R.M.; Spencer, H.H.; Patno, M.E. 
Cancer. 23, 1280 (1969). 

12. Smith, G.M.R.; Gordon, J. A.; Sewell , I. A.; Ellis, H. Br. J. 
Cancer. 21, 295 (1967). 

13. Nissen, N.I.; Pajak, T. ; Glidewell, 0.; Blom, H.; Flaherty, M. ; 
Hayes, D.; Mclntyre, R.; Holland, J.F. Cancer Treatment Rep. 
61, 1097 (1977). 

14. Farcier, R.J.; Mclntyre, O.R., Nissen, N.I.; Pajak, T.F.; 
Glidewell, 0.; Holland, J.F. Med. and Pediatr. Oncol. 4, 351 
(1978). 

15. Banzet, P.; Jacquellat, C. ; Civatte, J.; Puissant, A.; Moral, 
J.; Chastang, C; Israel, L; Belaich, S.; Jourdain, J.C.; Weil, 
M.; Auclerc, G. Cancer. 41, 1240 (1978). 

132 



133 



16. Cone, R.; Hasan, S.K.; Lown, J.W.; Morgan, A.R. Can. J. 
Biochem. 54, 219 (1976). 

17. Miller, D.S.; Laszlo, J.; McCarty, K.S.; Guild, W.R. ; 
Hochstein, P. Cancer Res. 27, 632 (1967). 

18. Rao, K.V. J. Pharm. Sci. 68, 853 (1979). 

19. White, H.L.; White, J.R. Mol . Pharmacol. 4, 549 (1968). 

20. Lown, J.W. Mol. and Cellular Biochem. 55, 17 (1983). 

21. Mizuno, N.S.; Gilboe, D.P. Biochim. Biophys. Acta 224, 319 
(1970). 

22. Kremer, W.B.; Laszlo, J. Hanbuch der experimentellen 
Pharmakologie 38, 633 (1975). 

23. Gollmain, A.P.; Takeshite, M. Adv. Enzyme. Regul . .18, 67 
(1980). 

24. Lown, J.W.; Sim, S.K. Can. J. Chem. 54, 2563 (1976). 

25. Hajdu, J.; Armstrong, E.C. J. Am. Chem. Soc. JL03, 232 (1981). 

26. Sinha, B.K. Chem. Biol. Interactions 36, 179 (1981). 

27. Miyasaka, T.; Hibino, S. ; Inouye, Y.; Nakamura, S. J. Chem. 
Soc. Perkin. Trans. I. 1986 , 479 (1986). 

28. Okada, H.; Inouye, Y.; Nakamura, S. J. Antibiotics 40, 230 
(1987). 

29. Inouye, Y.; Take, Y.; Oogose, K. ; Kubo, T. ; Nakamura, S. J. 
Antibiotics 40, 105 (1987). 

30. Inouye, Y.; Oogose, K.; Take, Y.; Kubo, A.; Nakamura, S. J. 
Antibiotics 40, 702 (1987). 

31. Take, Y.; Oogose, K.; Kubo, T. ; Inouye, Y.; Nakamura, S. J. 
Antibiotics 40, 679 (1987). 

32. Basha, F.Z.; Hibino, S. ; Kim, E. ; Pye, W.E.; Wu, T.T.; Weinreb, 
S.M. J. Am. Chem. Soc. 102, 3962 (1980). 

33. Kende, A.S.; Lorah, D.P.; Boatman, R.J. J. Am. Chem. Soc. 103, 
1271 (1981). 

34. Boger, D.L.; Panek, J.S. J. Org. Chem. 48, 623 (1983). 



134 

35. Rao, K.V. J. Heterocyclic Chem. 12, 725 (1975). 

36. Rao, K.V.; Venkateswarlu, P. J. Heterocyclic Chem. _12, 731 
(1975). 

37. Rao, K.V. J. Heterocyclic Chem. 14, 653 (1977). 

38. Liao, T.K.; Nyberg, W.H.; Cheng, C.C. J. Heterocyclic Chem. 
13, 1063 (1976). 

39. Liao, T.K.; Nyberg, W.H.; Cheng, C.C. Angew. Chem. Int. Ed. 6_, 
82 (1967). 

40. Hibino, S. ; Weinreb, S.M. J. Org. Chem. 42, 232 (1977). 

41. Kametani, T.; Kozuka, A.; Tanaka, S. Yakugaku Zasshi 90, 1574 
(1970). 

42. Liao, T.K.; Wittek, P.J.; Cheng, C.C. J. Heterocyclic Chem. 
_13, 1283 (1976). 

43. Rao, K.V. Cancer Chemotherapy Rep. Part 2 2, 11 (1974). 

44. Boger, D.L.; Yasuda, M. ; Mitscher, L.A.; Drake, S.D.; Kitos, 
P. A.; Thompson, S.C. J. Med. Chem. 30, 1918 (1987). 

45. Shaikh, I. A.; Johnson, F.; Grollman, A.P. J. Med. Chem. 29, 
1329 (1986). 

46. Baron, M. ; Giorgi-Renault, S. ; Mailliet, P.; Paoletti, C. ; 
Cros, S. Eur. J. Med. Chem. 18, 134 (1983). 

47. Renault, J.; Giorgi-Renault, S. ; Mailliet, P.; Baron, M. ; 
Paoletti, C; Cros, S. Eur. J. Med. Chem. 16, 24 (1981). 

48. Giorgi-Renault, S. ; Renault, J.; Baron, M. ; Servo! les, P.; 
Paoletti, C; Cros, S. Eur. J. Med. Chem. 20, 144 (1985). 

49. Parrot-Lopez, H. ; Delacotte, J.; Renault, J.; Cros, S. J. 
Heterocyclic Chem. 23, 1039 (1986). 

50. Delacotte, J.; Parrot-Lopez, H. ; Renault, J.; Cros, S. J. 
Heterocyclic Chem. 24, 571 (1987). 

51. Acheson, R.M. An Introduction to the Chemistry of Heterocyclic 
Compounds , Interscience Publishers, Inc., New York, 1960. 

52. Joseph, P.K.; Joullie, M.M. J. Med. Chem. I, 801 (1964). 



135 



53. Lora-Tamayo, M. ; Madronuo, R. ; Stud, M. Chem. Ber. 95, 2176 
(1962). 

54. Kubo, A.; Nakahara, S. ; Iwata, R. ; Takahashi , R. ; Arai , T. 
Tetrahedron Lett. 1980, 3207 (1980). 

55. Mclntyre, D.E.; Faulkner, D.J.; Van Engen, D.; Clardy, J. 
Tetrahedron Lett. 1979 , 4163 (1979). 

56. Kubo, A.; Nakahara, S. Chem. Pharm. Bull. 29, 595 (1981). 

57. Wolman, Y.; Gallop, P.M.; Patchornik, A.; Berger, A. J. Am. 
Chem. Soc. 89, 1889 (1962). 

58. Rao, K.V. Cancer Chemotherapy Rep. Part 2 2, 11 (1974). 

59. Prudhommeaux, E. ; Ermouf, G.; Foussard-Blampine, 0.; Viel, C. 
Eur. J. Med. Chem. 10, 19 (1975). 

60. Merz, K.W. ; Fink, J. Arch. Pharm. 289, 347 (1956). 

61. Boger, D.L.; Duff, S.R.; Panek, J.S.; Yasuda, M. J. Org. Chem. 
50, 5782 (1985). 

62. Atwell, G.J.; Rewcastle, G.W. ; Baguley, B.C.; Denny, W.A. J. 
Med. Chem. 30, 652 (1987). 

63. Iyengar, B.S.; Remers, W.A. ; Bradner, W.T. J. Med. Chem. 29, 
1864 (1986). 

64. Iyengar, B.S.; Sami , S.M.; Takahashi, T. ; Sikorskie, E.E.; 
Remers, W.A. ; Bradner, W.T. J. Med. Chem. 29, 1760 (1986). 

65. Noel, A.M.; Ryznerski, Z.; Berge, G.; Fulcrard, P.; Chevaliet, 
P.; Castel , J.; Orzalesi, H. Eur. J. Med. Chem. JL4, 135 
(1979). 

66. Finney, D.J. Probit Analysis , 3rd Edition, Cambridge 
University Press, New York, 1971. 

67. Strike, P.W. Medical Laboratory Statistics , Wright PSG, 
Littleton, MA, 1981. 



BIOGRAPHICAL SKETCH 
The author was born on December 28, 1956, in Winnsboro, SC, to 
Charles J. Jr. and Dorothy Bailey Beach, and grew up in Winnsboro. 
He attended Erskine College, Due West, SC, and received the Bachelor 
of Science degree in chemistry in 1979. He fulfilled his childhood 
goal of becoming a pharmacist in 1982, after graduating from The 
Medical University of South Carolina in Charleston, SC, with the 
Bachelor of Science in Pharmacy degree. With encouragement from the 
faculty there he entered the graduate program at the University of 
Florida, College of Pharmacy. His doctoral research was carried out 
under the supervision of Dr. K. V. Rao. 



136 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is 
fully adequate, in scope and quality, as a dissertation for the 
degree of Doctor of Philosophy. 




V.f^LQ 



Koppaka V. Ra6, Chairman 
Professor of Medicinal Chemistry 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is 
fully adequate, in scope and quality, as a dissertation for the 
degree of Doctor of Philosophy. 




Kenneth B. Sloan 
Associate Professor of Medicinal 
Chemistry 



I certify that I have read this study and that in my opinion it 
conforms to acceptable standards of scholarly presentation and is 
fully adequate, in scope and quality, as a dissertation for the 
degree of Doctor of Philosophy. 




a. 




n A. Zoltewicz 
ofessor of Chemistry 



This dissertation was submitted to the Graduate Faculty of the 
College of Pharmacy and to the Graduate School and was accepted as 
partial fulfillment of the requirements for the degree of Doctor of 
Philosophy. 

December 1987 




Dean, Graduate School